阅读说明：本技术 基于自然语言处理的处方生成方法、装置、设备及介质 (Prescription generating method, device, equipment and medium based on natural language processing ) 是由 陈莹 于 2021-09-08 设计创作，主要内容包括：本发明涉及人工智能技术领域,提出一种基于自然语言处理的处方生成方法、装置、设备及介质,该方法通过获取药品加购信息、用户描述信息分别生成适应症、可选使用规范信息、疑似症状和建议使用规范信息,进而生成处方必要信息并确定信息可信程度,得到初始处方。本发明还提出一种处方生成装置、设备及介质,借助于自然语言处理技术将用户的语言或文字描述信息以及药品加购信息转化为适应症、可选使用规范信息、疑似症状和建议使用规范信息,确定处方必要信息并生成初始处方,对于明确知晓自己症状及所需药品的患者,自动生成初始处方,减少患者的购药用时,提升患者的体验度和满意度,也提升了医生的接诊效率和医疗资源利用率。(The invention relates to the technical field of artificial intelligence, and provides a prescription generation method, a device, equipment and a medium based on natural language processing. The invention also provides a prescription generating device, equipment and a medium, which convert the language or text description information and the medicine purchase information of the user into indication, optional use specification information, suspected symptoms and recommended use specification information by means of a natural language processing technology, determine the necessary information of the prescription and generate the initial prescription, automatically generate the initial prescription for the patient who definitely knows own symptoms and the required medicine, reduce the medicine purchase of the patient, improve the experience and satisfaction of the patient, and improve the doctor's treatment efficiency and the medical resource utilization rate.)

1. A natural language processing based prescription generating method, the method comprising:

acquiring medicine purchase information, and determining indications and optional use standard information of purchased medicines;

acquiring user description information and determining main complaint information;

generating necessary prescription information according to the main complaint information and the medicine purchase information, wherein the necessary prescription information comprises at least one of suspected symptoms and recommended use standard information;

determining the information credibility of the necessary prescription information according to at least one of symptom information and specification information, wherein the symptom information is determined according to the indication and suspected symptoms, and the specification information is determined according to the optional use specification information and recommended use specification information;

and generating an initial prescription according to the information credibility and the necessary information of the prescription.

2. The prescription generating method of claim 1, wherein said at least one of symptom information and normative information determining an information credibility of said prescription required information comprises:

determining the credibility of the symptoms according to the indications and the suspected symptoms;

determining the credibility of the use specification according to the optional use specification information and the suggested use specification information;

if the symptom credibility and the use specification credibility are both credible, the information credibility is credible;

and if any one of the symptom credibility and the use specification credibility is not credible, the information credibility is not credible.

3. The prescription generating method of claim 1, wherein said generating an initial prescription based on said information credibility and prescription required information comprises any one of:

if the information credibility degree comprises credibility, generating the initial prescription according to the necessary information of the prescription;

or if the information credibility degree comprises unreliability, acquiring user associated information, determining the importance degree of a purchase adding event according to at least one of the user associated information, user description information and medicine purchase adding information, wherein the user associated information comprises at least one of a user position, a user purchase frequency and a historical transaction amount, distributing a doctor receiving according to the importance degree of the purchase adding event, sending the necessary prescription information to the doctor receiving, acquiring a prescription instruction of the doctor receiving and generating the initial prescription.

4. The prescription generating method of claim 3,

determining the importance degree of the purchase adding event according to the user description information comprises the steps of obtaining emotional information in the user description information, determining the emotional state of the user and determining the importance degree of the purchase adding event, wherein the emotional information comprises at least one of expression, action, tone, speed and language word;

the method comprises the steps that the medicine purchase information comprises purchase amount and waiting time, the importance degree of the purchase event is determined according to the medicine purchase information, and the importance degree of the purchase event is determined according to the purchase amount and the waiting time.

5. The prescription generating method of any one of claims 1 to 4, wherein said obtaining user description information and determining the complaint information comprises:

determining symptom questions according to the indications, inquiring for the first time, and obtaining a first answer result;

inputting the answer result into a preset keyword extraction model to obtain initial information;

generating a confirmation question according to the initial information, and carrying out a second inquiry to obtain a second answer result;

and if the second answer result comprises the confirmation of the initial information, generating the chief complaint information according to the initial information.

6. The prescription generating method of any of the claims 1-4, wherein after generating an initial prescription based on said information credibility and prescription required information, said method further comprises,

verifying the initial prescription;

if the verification is passed, generating a formal prescription according to the initial prescription;

acquiring the prescription quantity, the inventory quantity and the preset limit purchase quantity of prescription medicines in the formal prescription, and generating a medicine sales list;

the generation mode of the drug sales list comprises any one of the following modes:

if the prescription quantity is not greater than the preset limit purchase quantity and the prescription quantity is not greater than the inventory quantity, the medicine sales list comprises the prescription medicine and the prescription quantity;

if the stock quantity is greater than 0, the prescription quantity is not greater than the preset limited purchase quantity, and the stock quantity is less than the prescription quantity, the medicine sales list comprises the prescription medicine and the stock quantity;

if the stock quantity is equal to 0, carrying out the shortage prompt of the prescription medicine on a medicine sales list;

or if the prescription quantity is not greater than the inventory quantity, the prescription quantity is greater than a preset limit purchase quantity, the medicine sales list comprises the prescription medicines and the preset limit purchase quantity, and a prescription quantity reduction prompt is performed.

7. The prescription generating method of claim 5, wherein before the acquiring the prescription amount, the inventory amount and the preset limit purchase amount of the prescription drugs in the official prescription and generating the drug sales list, the method further comprises:

acquiring historical order information and current user identification information, wherein the current user identification information comprises at least one of current purchase adding user identity information and current payment identification information, and the historical order information comprises at least one of historical purchase adding user identity information, historical purchase adding medicines, historical purchase adding medicine quantity and historical payment identification information;

determining a user transaction frequency according to the historical order information and the current user identification information, wherein the user transaction frequency comprises at least one of the current purchase adding user identity information transaction frequency and the current payment identification information transaction frequency;

if the user transaction frequency is higher than a preset transaction frequency threshold value, suspending the transaction;

if the user transaction frequency is lower than a preset transaction frequency threshold value, determining the user key medicine transaction quantity according to the historical order information and the current user identification information, wherein the user key medicine transaction quantity comprises at least one of the current purchased user identity information key medicine transaction quantity and the current payment identification information key medicine transaction quantity, and if the user key medicine transaction quantity exceeds the preset transaction quantity threshold value, suspending the transaction;

or if the transaction frequency of the user is lower than a preset transaction frequency threshold value and the transaction quantity of the key medicines of the user is lower than a preset transaction quantity threshold value, continuing the transaction.

8. A natural language processing based prescription generating apparatus, the apparatus comprising:

the purchase information acquisition module is used for acquiring medicine purchase information and determining indications and optional use standard information of purchased medicines;

the description information acquisition module is used for acquiring user description information and determining the chief complaint information;

the prescription necessary information generating module is used for generating prescription necessary information according to the chief complaint information and the medicine purchase information, and the prescription necessary information comprises at least one of suspected symptoms and recommended use standard information;

the determining module is used for determining the information credibility of the necessary information of the prescription according to at least one of symptom information and specification information, wherein the symptom information is determined according to the indication and suspected symptoms, and the specification information is determined according to the optional use specification information and the recommended use specification information;

and the initial prescription generating module is used for generating an initial prescription according to the information credibility and the necessary prescription information.

9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 7 are implemented by the processor when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.

Technical Field

The invention relates to the technical field of artificial intelligence, and provides a natural language processing-based prescription generation method, device, equipment and medium.

Background

In the related art, the prescription process generally includes a doctor and a patient to perform a dialogue, and after confirming patient information, the doctor fills in diagnosis, prescribes a medicine, and signs, and finally obtains a prescription.

However, the existing inquiry flow needs to be inquired by a doctor no matter whether a patient clearly knows the required medicine, so that the time consumption of a part of patients who clearly know the condition of the patient and the required medicine is long, the experience degree of the patient is reduced, and the satisfaction degree is poor. In addition, for diagnosis and treatment resources, if each patient needs to be subjected to in-person inquiry, medical resources are wasted to a certain extent, and limited diagnosis and treatment resources are evenly distributed to some patients who do not need to be subjected to in-person inquiry, so that the treatment efficiency is low.

Disclosure of Invention

The invention provides a prescription generating method, a device, equipment and a medium based on natural language processing, and the method, the device, the equipment and the medium are mainly used for converting language or text description information and medicine purchase information of a user into indication, optional use specification information, suspected symptoms and recommended use specification information by means of a natural language processing technology, further determining necessary prescription information and information credibility thereof, generating an initial prescription based on the information credibility and the necessary prescription information, avoiding the need of a doctor for asking each patient to make an inquiry on the spot, generating the initial prescription in an automatic checking mode for the patient who definitely knows own symptoms and required medicines, and improving the experience degree and satisfaction degree of the patient when the patient purchases medicines, and also improving the diagnosis receiving efficiency and medical resource utilization rate of the doctor.

In order to achieve the above object, the present invention provides a recipe generating method based on natural language processing, the method including:

acquiring medicine purchase information, and determining indications and optional use standard information of purchased medicines;

acquiring user description information and determining main complaint information;

generating necessary prescription information according to the main complaint information and the medicine purchase information, wherein the necessary prescription information comprises at least one of suspected symptoms and recommended use standard information;

determining the information credibility of the necessary prescription information according to at least one of symptom information and specification information, wherein the symptom information is determined according to the indication and suspected symptoms, and the specification information is determined according to the optional use specification information and recommended use specification information;

and generating an initial prescription according to the information credibility and the necessary information of the prescription.

Optionally, the determining the information credibility of the information necessary for the prescription according to at least one of the symptom information and the specification information includes:

determining the credibility of the symptoms according to the indications and the suspected symptoms;

determining the credibility of the use specification according to the optional use specification information and the suggested use specification information;

if the symptom credibility and the use specification credibility are both credible, the information credibility is credible;

and if any one of the symptom credibility and the use specification credibility is not credible, the information credibility is not credible.

Optionally, the generating an initial prescription according to the information credibility and the necessary prescription information includes any one of the following:

if the information credibility degree comprises credibility, generating the initial prescription according to the necessary information of the prescription;

or if the information credibility degree comprises unreliability, acquiring user associated information, determining the importance degree of a purchase adding event according to at least one of the user associated information, user description information and medicine purchase adding information, wherein the user associated information comprises at least one of a user position, a user purchase frequency and a historical transaction amount, distributing a doctor receiving according to the importance degree of the purchase adding event, sending the necessary prescription information to the doctor receiving, acquiring a prescription instruction of the doctor receiving and generating the initial prescription.

Optionally, determining the importance degree of the purchase adding event according to the user description information includes acquiring emotion information in the user description information, determining the emotional state of the user, and determining the importance degree of the purchase adding event, where the emotion information includes at least one of expression, action, tone, speed of speech, and word of tone;

the method comprises the steps that the medicine purchase information comprises purchase amount and waiting time, the importance degree of the purchase event is determined according to the medicine purchase information, and the importance degree of the purchase event is determined according to the purchase amount and the waiting time.

Optionally, the obtaining user description information and determining the complaint information include:

determining symptom questions according to the indications, inquiring for the first time, and obtaining a first answer result;

inputting the answer result into a preset keyword extraction model to obtain initial information;

generating a confirmation question according to the initial information, and carrying out a second inquiry to obtain a second answer result;

and if the second answer result comprises the confirmation of the initial information, generating the chief complaint information according to the initial information.

Optionally, after generating the initial prescription according to the information credibility and the information necessary for the prescription, the method further comprises,

verifying the initial prescription;

if the verification is passed, generating a formal prescription according to the initial prescription;

acquiring the prescription quantity, the inventory quantity and the preset limit purchase quantity of prescription medicines in the formal prescription, and generating a medicine sales list;

the generation mode of the drug sales list comprises any one of the following modes:

if the prescription quantity is not greater than the preset limit purchase quantity and the prescription quantity is not greater than the inventory quantity, the medicine sales list comprises the prescription medicine and the prescription quantity;

if the stock quantity is greater than 0, the prescription quantity is not greater than the preset limited purchase quantity, and the stock quantity is less than the prescription quantity, the medicine sales list comprises the prescription medicine and the stock quantity;

if the stock quantity is equal to 0, carrying out the shortage prompt of the prescription medicine on a medicine sales list;

or if the prescription quantity is not greater than the inventory quantity, the prescription quantity is greater than a preset limit purchase quantity, the medicine sales list comprises the prescription medicines and the preset limit purchase quantity, and a prescription quantity reduction prompt is performed.

Optionally, before acquiring the prescription quantity, the inventory quantity, and the preset limit purchase quantity of the prescription drugs in the official prescription and generating the drug sales list, the method further includes:

acquiring historical order information and current user identification information, wherein the current user identification information comprises at least one of current purchase adding user identity information and current payment identification information, and the historical order information comprises at least one of historical purchase adding user identity information, historical purchase adding medicines, historical purchase adding medicine quantity and historical payment identification information;

determining a user transaction frequency according to the historical order information and the current user identification information, wherein the user transaction frequency comprises at least one of the current purchase adding user identity information transaction frequency and the current payment identification information transaction frequency;

if the user transaction frequency is higher than a preset transaction frequency threshold value, suspending the transaction;

if the user transaction frequency is lower than a preset transaction frequency threshold value, determining the user key medicine transaction quantity according to the historical order information and the current user identification information, wherein the user key medicine transaction quantity comprises at least one of the current purchased user identity information key medicine transaction quantity and the current payment identification information key medicine transaction quantity, and if the user key medicine transaction quantity exceeds the preset transaction quantity threshold value, suspending the transaction;

or if the transaction frequency of the user is lower than a preset transaction frequency threshold value and the transaction quantity of the key medicines of the user is lower than a preset transaction quantity threshold value, continuing the transaction.

In addition, to achieve the above object, the present invention provides a prescription generating apparatus based on natural language processing, the apparatus including:

the purchase information acquisition module is used for acquiring medicine purchase information and determining indications and optional use standard information of purchased medicines;

the description information acquisition module is used for acquiring user description information and determining the chief complaint information;

the prescription necessary information generating module is used for generating prescription necessary information according to the chief complaint information and the medicine purchase information, and the prescription necessary information comprises at least one of suspected symptoms and recommended use standard information;

the determining module is used for determining the information credibility of the necessary information of the prescription according to at least one of symptom information and specification information, wherein the symptom information is determined according to the indication and suspected symptoms, and the specification information is determined according to the optional use specification information and the recommended use specification information;

and the initial prescription generating module is used for generating an initial prescription according to the information credibility and the necessary prescription information.

Furthermore, to achieve the above object, the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the method according to any one of the above embodiments.

Furthermore, to achieve the above object, the present invention further provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the method according to any one of the above embodiments.

The invention provides a prescription generating method, a device, equipment and a medium based on natural language processing, the method respectively generates indication, optional use specification information, suspected symptom and recommended use specification information by acquiring medicine purchase information and user description information, generates prescription necessary information and information credibility thereof, further generates an initial prescription, converts the language or character description information and the medicine purchase information of a user into the indication, optional use specification information, suspected symptom and recommended use specification information by means of natural language processing technology, can effectively reduce the judgment time of whether the purchase medicine is suitable for the user, can generate the initial prescription in an automatic checking mode by determining the prescription necessary information and information credibility if the user clearly knows own symptoms and required medicines and determining the information credibility including credibility according to the data, the experience degree and the satisfaction degree of the patient can be improved when the patient purchases the medicine, if the credibility degree of the information includes the unreliability, the necessary information of the prescription can be provided for the corresponding doctor to receive a doctor, the doctor-patient communication time is saved, and the experience degree and the satisfaction degree of the patient can be improved when the patient purchases the medicine.

Drawings

FIG. 1 is a schematic flow chart of a recipe generation method based on natural language processing according to an embodiment of the present invention;

FIG. 2 is another flow diagram of recipe generation based on natural language processing provided in an embodiment of the present invention;

FIG. 3 is another flow diagram of recipe generation based on natural language processing provided in an embodiment of the present invention;

FIG. 4 is a flowchart illustrating a specific process for recipe generation based on natural language processing according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a natural language processing based prescription generating apparatus according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a computer device provided in an embodiment of the present invention.

The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

In one embodiment, a natural language processing based recipe generation method is provided, as shown in fig. 1, the method comprising the steps of:

step S101: acquiring medicine purchase information, and determining indications and optional use standard information of purchased medicines.

Optionally, the medicine purchase adding information may be determined according to a purchase adding action, a purchase adding result, and the like of the user. The medicine purchase information includes, but is not limited to, the name of the medicine to be purchased, the barcode of the medicine to be purchased, the quantity of the medicine to be purchased, the price of the medicine to be purchased, the effective components of the medicine to be purchased, and the like.

Optionally, the optional use specification information includes, but is not limited to, optional dosing frequency, optional amount, and optional usage. The administration frequency is once a day, twice a day, etc., the dosage can be selected from 1 granule once, 1-2 granules once, etc., and the dosage can be selected from oral administration, external application, etc.

In some embodiments, the manner in which indications and optional use specification information are determined from the drug purchase information includes at least one of:

acquiring a specification of each purchased medicine, and determining indication and optional use standard information according to description information about the indication in the specification;

determining indication and optional use standard information through a preset medicine information platform according to the medicine purchase information;

and determining the indications and the optional use standard information through the preset medicine indication mapping relation.

Optionally, the relevant data of the instruction book may be stored in one or more servers or local databases, and the instruction book corresponding to the acquired purchased medicine may be determined by extracting the description information about the indication part in the instruction book according to the barcode or other unique identification information of the purchased medicine.

Optionally, the instruction book may also be searched by a search engine or the like based on the location identity information of the purchased medicine to obtain the instruction book, and further obtain the indication and the optional use specification information.

Since indication and optional use specification information (such as using method, dosage, frequency and the like) in the specification of the medicine are necessarily existing information, identification and key information extraction can be directly carried out on the basis of the specification to obtain the indication and optional use specification information. For example, full text of the specification can be input into the system, and based on keyword search, the corresponding description position of the specification is located to obtain indication and optional use specification information.

Optionally, the preset mapping relationship of the drug indications can be marked and formulated by professionals such as doctors and pharmacists according to the indications of a certain drug and the information of the optional application range of the tablet, and can also be formulated by other ways known by those skilled in the art.

The preset drug information platform may be a platform known to those skilled in the art, and is not limited herein.

Alternatively, the indication of indication or suspected symptom can be realized by using ICD (international classification of diseases) 10 standard code.

In some embodiments, after step S101, before step S102, the method further comprises:

determining purchased medicines and the quantity of the purchased medicines according to the medicine purchase information;

acquiring the quantity of the stocked medicines for purchasing and presetting the maximum purchasing quantity;

and determining the reasonable medicine quantity according to the quantity of the stored medicines, the preset maximum purchase quantity and the purchase medicine quantity, and updating the medicine purchase information according to the reasonable medicine quantity.

The preset maximum purchase amount can be set by those skilled in the art as desired. The quantity of the stored medicines can be obtained from the inventory statistical information of a pharmacy or a medicine selling platform.

Optionally, the reasonable quantity of the medicines is determined according to the quantity of the stocked medicines, the preset maximum purchase quantity and the purchase quantity of the medicines, and the medicine purchase information is updated according to the reasonable quantity of the medicines, wherein the medicine purchase information comprises:

if the number of purchased medicines is not greater than the preset maximum purchase amount and the number of purchased medicines is not greater than the number of inventory medicines, executing step S102;

if the quantity of the stored medicines is 0, deleting the purchased medicines and the quantity of the purchased medicines in the medicine purchasing information, and prompting that the purchased medicines are out of stock;

if the quantity of the stocked medicines is larger than 0, the quantity of the purchased medicines is not larger than the preset maximum purchased quantity, and the quantity of the purchased medicines is larger than the quantity of the stocked medicines, the quantity of the purchased medicines in the medicine purchased information is modified into the quantity of the stocked medicines, and the quantity of the purchased medicines is prompted to be modified.

Optionally, if the quantity of the inventory drugs is 0, the method further includes:

and recommending the substitute medicines for the user according to a preset recommendation rule.

The preset recommendation rule comprises a plurality of preset groups of medicine combinations which can be replaced mutually.

Optionally, if the user agrees to replace the purchased medicine with the substitute medicine, the step S102 is continuously executed, otherwise, the step is stopped.

It should be noted that, the above quantity of inventory drugs being 0 may be understood as a situation where there is no drug in the current inventory, and does not refer to that the inventory of the drug is only 0, for example, due to an inventory management capability problem, there may be a "virtual inventory" in the quantity of inventory, and the inventory data may be 1 or more at this time, but actually there is no real goods in the warehouse, and the inventory in this case may be regarded as 0 at this time.

In some embodiments, when the user performs the purchase adding operation, the inventory check may be performed, and when the inventory is smaller than the purchase adding amount, the purchase adding is prompted to fail, and the purchase adding is rejected.

Through the implementation mode, the requirements of a user on medicines and diseases which the user may need to treat can be known, meanwhile, the current inventory condition and the quantity of the medicines required by the user can be preliminarily compared, if the inventory cannot meet the requirements of the user and is informed to the user in time, the recommendation of alternative medicines is provided on the feasible premise, for example, medicines of different manufacturers with the same effective component or medicines with the same effective component and different dosage specifications are recommended, and compared with the case that the user is informed of the shortage of certain medicines after the prescription audit is completed, the experience degree of the user can be further improved, and the problem that the user cannot take the medicines for symptoms in time due to shortage of the medicines is avoided.

In this embodiment, the source of the user description information may be description information of a person who actually uses the additional medicine, and the person who actually uses the additional medicine may be a person who performs the additional medicine purchasing operation, or may be another person.

S102: and acquiring user description information and determining the main complaint information.

Optionally, the obtaining manner of the user description information includes, but is not limited to, at least one of the following:

the user inputs information, for example, the user directly performs information oral or filling input;

obtaining the associated information of the user login account, for example, if the user logs in through a medical insurance card, an account and the like, obtaining the basic information of the user based on the associated information of the account;

based on the answer to the question, etc.

Optionally, basic user information can be determined according to the user description information, the basic user information includes but is not limited to user name, age, native place, address, past medical history and the like, and the chief complaint information includes but is not limited to symptoms, duration and the like.

In some embodiments, referring to fig. 2, obtaining the user description information and determining the complaint information includes:

s201: determining symptom questions according to the indications, inquiring for the first time, and obtaining a first answer result;

s202: inputting the answer result into a preset keyword extraction model to obtain initial information;

s203: generating a confirmation question according to the initial information, and inquiring for the second time to obtain a second answer result;

s204: and if the second answer result comprises the confirmed initial information, generating the main complaint information according to the initial information.

Optionally, the user basic information may be generated according to the initial information.

The corresponding preset symptom questions can be set for each indication in advance by a professional to form a preset symptom question set, the corresponding symptom questions can be determined according to the indication, and the first inquiry is carried out according to the symptom questions to obtain a first answer result. A plurality of symptom questions are determined based on the indications, and questions can be specifically asked according to the symptoms of diseases required to be treated by the user, so that whether the user possibly has the symptoms which can be treated by the purchased medicines can be judged according to the answer results.

Optionally, the first query may be a query process including a plurality of symptom questions, and in this case, the process may be: firstly, finding symptom questions in preset symptom questions according to indications, inquiring to obtain an answer result 1, then determining supplementary questions according to the answer result, performing supplementary inquiry to obtain an answer result 2, and further synthesizing the answer result 1 and the answer result 2 to generate a first answer result. The supplementary questions can be determined by predicting various answer results 1 in advance and setting a plurality of supplementary question sets correspondingly. Optionally, the supplementary questions may be a plurality of questions preset by those skilled in the art according to needs, a plurality of questions randomly selected from a preset question library and different from the symptom questions, or a preset associated question according to at least one question in the symptom questions.

Optionally, the process of querying based on the symptom questions and the supplemental questions comprises any one of:

alternating symptomatic and supplemental questions;

asking first a symptom question and then a supplemental question;

asking supplemental questions first and then symptom questions;

and randomly selecting symptom questions or supplementary questions to inquire.

It should be noted that the query may be implemented by voice and/or text, and the corresponding answer result may also be implemented by voice and/or text.

The preset keyword extraction model may be a model pre-trained in a manner known to those skilled in the art, and is not limited herein. The preset keyword extraction model includes a segmentation model and a keyword extraction model, for example, the segmentation model constructed according to a preset segmentation rule performs segmentation processing on the answer result to obtain a plurality of segmentation, the segmentation is input into the extraction model, cosine similarity between segmentation vectors of each segmentation and keyword vectors of a plurality of preset keywords is respectively obtained, and then keywords corresponding to the segmentation are obtained to obtain initial information.

Optionally, the answer result may be subjected to word segmentation processing to obtain a plurality of segmented words, then the obtained segmented words are compared with the initial words in the preset word bank according to word meaning similarity, and the initial word which has the highest similarity with a certain segmented word and is higher than a preset word meaning similarity threshold is used as the initial information of the segmented word.

By the method, on one hand, the spoken expressions of the user can be converted into professional expressions, and on the other hand, the accuracy of the initial information extracted by the preset keyword extraction model can be ensured through the answers of the user to the confirmation questions, so that the accuracy of the follow-up prescription is ensured.

Optionally, in the process of generating the confirmation question according to the initial information, the confirmation question may be directly added to the initial information through a preset confirmation question template. Sometimes, the initial information may be over-written and expressed, so that some users may not understand the meaning of the initial information, or the initial information may be spoken information obtained by performing spoken processing on the initial information, and then the spoken information is added to a preset confirmation question template to generate a confirmation question, so that the user can clearly understand the question and make a correct answer. For example, the confirmation question template is preset as follows: ask you for what means + initial information + do? Assuming that the initial information is "acne", the confirmation question obtained according to the first method is "ask you for acne, and the confirmation question obtained according to the second method is obtained by colloquially treating acne to obtain pox, and further obtaining the confirmation question" ask you for pox.

S103: and generating necessary prescription information according to the main complaint information and the medicine purchase information.

Optionally, the prescription necessary information includes at least one of suspected symptoms and recommended use specification information.

Optionally, the generating of the necessary prescription information according to the complaint information and the purchased medicine information includes:

comparing the main complaint information with a preset necessary information set to obtain a preset necessary information with the highest information similarity between the main complaint information and the preset necessary information set;

and if the information similarity of the preset necessary information is higher than a preset information similarity threshold, taking the preset necessary information as the prescription necessary information of the chief complaint information.

Specifically, the information similarity may be determined by determining a first word vector of each word in the chief complaint information, obtaining a second word vector of each word in each preset necessary information, and determining cosine similarity between the first word vector and the second word vector as the information similarity.

The preset necessary information comprises at least one of suspected symptoms and recommended use specification information, and the recommended use specification information comprises at least one of recommended administration frequency, recommended dosage and recommended usage.

Through the implementation mode, the credible necessary information of the prescription can be converted based on the user dictation or the input irregular user description information, so that the subsequent prescription can be accurately generated, and the recommended usage dosage administration frequency of the medicine purchased by the user is given.

Optionally, the necessary information of the prescription may include basic information of the user in addition to the suspected symptom and the recommended use specification information, and each data may be filled in a preset form for extracting the subsequent information.

S104: the information credibility of the prescription necessary information is determined based on at least one of the symptom information and the specification information.

Wherein, symptom information is determined according to indications and suspected symptoms, and specification information is determined according to optional use specification information and recommended use specification information.

In some embodiments, determining the information credibility of the prescription required information based on at least one of the symptom information and the specification information comprises:

determining the credibility of the symptoms according to the indications and the suspected symptoms;

determining the credibility of the use specification according to the optional use specification information and the suggested use specification information;

if the symptom credibility and the use specification credibility are both credible, the information credibility is credible;

if any one of the symptom credibility and the use specification credibility is not credible, the information credibility is not credible.

Wherein if the indication comprises suspected symptoms, the credible condition of the symptoms comprises credibility, otherwise, the credibility is not; if the optional use specification information comprises the recommended use specification information, the use specification credibility condition comprises credibility, otherwise, the use specification credibility is not credible. When the suspected symptoms are empty, that is, reasonable suspected symptoms are not obtained according to the chief complaint information, the credible condition of the symptoms comprises unreliability; when the recommended use specification information is empty, that is, no reasonable recommended use specification information is obtained according to the purchased medicine information, the credibility condition of the use specification includes unreliability.

The above "including" may be the same or partially the same, and for example, the indication includes "antipyretic analgesic". The medicine is used for treating headache, fever, nasal obstruction, watery nasal discharge and pharyngalgia caused by cold, and the suspected symptoms are nasal obstruction caused by cold, and the indication disease comprises the suspected symptoms. For another example, the optional use specification information may include "one-piece-a-day, oral-administration", the recommended use specification information may include "one-piece-a-day, oral-administration", the optional use specification information may include the recommended use specification information, and the use specification trust status may include trust. For another example, the optional use specification information includes "1-2 tablets per day for oral administration", the recommended use specification information includes "1 tablet per day for oral administration", the optional use specification information includes recommended use specification information, and the use specification credibility includes credibility.

S105: and generating an initial prescription according to the information credibility and the necessary information of the prescription.

In some embodiments, generating the initial prescription based on the information credibility and the prescription requisite information includes any one of:

if the information credibility degree includes credibility, generating an initial prescription according to the necessary information of the prescription;

or if the information credibility degree comprises unreliability, acquiring user associated information, determining the importance degree of the purchase adding event according to at least one of the user associated information, the user description information and the medicine purchase adding information, wherein the user associated information comprises at least one of the user position, the user purchase frequency and the historical transaction amount, allocating a doctor receiving according to the importance degree of the purchase adding event, sending necessary prescription information to the doctor receiving, acquiring a prescription instruction of the doctor receiving, and generating an initial prescription.

In some embodiments, determining the importance of the shopping events based on the user profile includes: obtaining emotional information in the user description information, determining the emotional state of the user, and determining the importance degree of the purchase adding event, wherein the emotional information comprises at least one of expression, action, tone, speed of speech and tone words.

The emotion identification model can be trained in advance to determine the emotion state of the user through emotion information. And carrying out quantifiable index grading on different emotional states according to a preset rule to obtain the importance degree of the purchase adding event. The emotion recognition model can be trained in a manner known to those skilled in the art.

In some embodiments, the drug purchase information includes a purchase amount and a wait period, and determining the importance of the purchase event according to the drug purchase information includes: and determining the importance degree of the purchase adding event according to the purchase adding amount and the waiting time.

Optionally, if the user association information includes the user position, determining a position attribute (hospital, market, house, school, etc.) according to the user position, presetting an influence parameter for different position attributes, and further evaluating the importance degree of the purchase adding event.

Through the parameter indexes which are quantized, a quantized purchasing event importance degree can be obtained, and comparison of a plurality of purchasing events according to the purchasing event importance degree is facilitated.

The importance degree of the medicine purchase adding event is comprehensively evaluated through various factors, and the importance degree can influence the subsequent processing speed of the event, so that part of special users can be better served.

For example, the user association information, the user description information, and the medicine purchase adding information may be respectively scored or graded, so as to obtain the importance degree of the purchase adding event. Specifically, a plurality of influence factors may be preset for evaluation items included in each of the user association information, the user description information, and the medicine purchase adding information, a quantitative value, such as purchase frequency, historical transaction amount, purchase adding amount, waiting time, and the like, may be selected for the items that can be quantified, and the quantitative values are multiplied by the influence factors respectively for summation. And giving a certain numerical value to unquantized parts in the evaluation items such as emotional information, user positions and the like in advance, multiplying and summing the numerical values by the influence factors respectively, and finally obtaining a total summation numerical value so as to measure the importance degree of the purchase adding event by the total summation numerical value.

In some embodiments, the manner of assigning the referring physician comprises:

and determining the department according to at least one of the indication and the suspected symptom, and determining a doctor who meets a doctor among the available doctors in the department.

Optionally, if the department determined by the indication is inconsistent with the department determined by the suspected symptom, the department determined by the suspected symptom may be referred to or assigned to a general practitioner.

By the aid of the method, the prescription or the manual prescription can be automatically generated according to the credibility of the necessary information of the prescription, the user can be effectively distributed, and the user who can be automatically issued directly through the system can be automatically issued directly in an automatic issuing mode, so that the workload of a doctor is effectively reduced, the working efficiency is improved, the waiting time of the user is reduced, and the satisfaction degree of the user is improved. Meanwhile, the doctor visits are sequenced according to the order of the purchase adding event importance degrees, so that users with partial demands being more urgent (in hospitals, splenic dysphoria and urgent demands) can be served preferentially, the occurrence of customer complaints can be avoided to a certain extent, and the users with urgent demands are met preferentially.

In some embodiments, after step S101, the method further comprises:

determining whether the purchased medicine in the medicine purchase information is the limited purchased medicine;

if yes, distributing the doctor of receiving a doctor and manually generating an initial prescription;

if not, go to step S102.

Optionally, the list of the limited purchased medicines is preset, and then whether the distribution doctor is assigned to manually prescribe the prescription can be determined by judging whether the purchased medicines are the medicines for limiting the purchase of the medicine list. The limit on purchasing a drug list may be determined by one skilled in the art as desired.

For example, some drugs for treating mental diseases have great side effects, some drugs have addiction, and doctors are required to judge whether to administer the drugs according to the actual conditions of users, and at the moment, the prescriptions are not suitable for being made in an automatic prescription making mode, so that corresponding doctor-taking doctors are allocated to the users for manual prescription making.

In some embodiments, referring to fig. 3, after generating the initial prescription based on the information credibility and the prescription required information, the method further comprises:

s301: checking the initial prescription;

s302: if the verification is passed, generating a formal prescription according to the initial prescription;

s303: acquiring the prescription quantity, the inventory quantity and the preset limit purchase quantity of the prescription medicines in the official prescription, and generating a medicine sales list.

The verification method includes, but is not limited to, verifying the initial prescription by a professional (pharmacist, etc.) through manual verification.

In one embodiment, the method for generating the drug sales inventory includes:

the method comprises the steps of obtaining prescription drugs and prescription quantity in an initial prescription, obtaining stock quantity and preset limit purchase quantity of the prescription drugs, and generating a drug sales list according to the prescription quantity, the stock quantity and the preset limit purchase quantity.

Optionally, the generation manner of the drug sales list includes any one of the following:

if the prescription quantity is not greater than the preset limit purchase quantity and the prescription quantity is not greater than the inventory quantity, the medicine sales list comprises prescription medicines and the prescription quantity;

if the stock quantity is greater than 0, the prescription quantity is not greater than the preset limit purchase quantity, and the stock quantity is less than the prescription quantity, the medicine sales list comprises prescription medicines and the stock quantity;

if the stock quantity is equal to 0, carrying out the shortage prompt of the prescription medicine on the medicine sales list;

and if the prescription quantity is not larger than the inventory quantity, the prescription quantity is larger than the preset limit purchase quantity, the medicine sales list comprises the prescription medicines and the preset limit purchase quantity, and the prescription quantity reduction prompt is carried out.

The predetermined limit purchase amount may be set according to various related regulations for different prescribed drugs.

Through the embodiment, whether the quantity of the stored medicines meets the requirements of the user can be further verified, whether the quantity of the medicines purchased by the user exceeds the limited purchase quantity can be verified, the reasonability of the quantity of the medicines listed on the opened medicine sales list can be ensured, and the problem that the user cannot timely take the medicines with the proper quantity due to shortage of the medicines is avoided.

Optionally, the out-of-stock prompt and the prescription quantity reduction prompt may be displayed to the user by voice, text, or the like. The specific manner of prompting is not limited herein.

In some embodiments, some drugs have large price differences between different regions due to local policy and other factors, and some people may purchase the drugs too frequently to make a stockpile and a sale for profit. Or, some people frequently perform the medicine purchasing action for various reasons, which results in the waste of resources. Therefore, before acquiring the prescription quantity, the inventory quantity and the preset limit purchase quantity of the prescription drugs in the formal prescription and generating the drug sales list, the method further comprises the following steps:

acquiring historical order information and current user identification information, wherein the current user identification information comprises at least one of current purchase adding user identity information and current payment identification information, and the historical order information comprises at least one of historical purchase adding user identity information, historical purchase adding medicines, historical purchase adding medicine quantity and historical payment identification information;

determining user transaction frequency according to historical order information and current user identification information, wherein the user transaction frequency comprises at least one of current purchase adding user identity information transaction frequency and current payment identification information transaction frequency;

if the transaction frequency of the user is higher than a preset transaction frequency threshold value, suspending the transaction;

if the user transaction frequency is lower than a preset transaction frequency threshold value, determining the user key medicine transaction quantity according to the historical order information and the current user identification information, wherein the user key medicine transaction quantity comprises at least one of the current purchase user identity information key medicine transaction quantity and the current payment identification information key medicine transaction quantity, and if the user key medicine transaction quantity exceeds the preset transaction quantity threshold value, suspending the transaction;

and if the transaction frequency of the user is lower than the preset transaction frequency threshold value and the transaction quantity of the key medicines of the user is lower than the preset transaction quantity threshold value, continuing the transaction.

The current purchase adding user identity information may be user identity information for executing purchase adding actions, such as a user name, a user ID and other user unique values, and the payment identification information may be a payment account and the like.

The preset transaction frequency threshold and the preset transaction quantity threshold can be set by those skilled in the art according to needs.

The current mode for determining the transaction frequency of the identity information of the additional purchase user can be as follows:

obtaining historical order information in a certain time period, wherein the historical order information comprises a transaction order (placed and paid), an unpaid order (placed and unpaid), and a purchase order (not placed and unpaid and purchased), and determining the transaction frequency of the identity information of the current purchase user by counting the occurrence amount of the identity information of the current purchase user in the historical order information.

The determination mode of the transaction frequency of the current payment identification information comprises the following steps:

obtaining historical order information in a certain time period, wherein the historical order information comprises a transaction order (placed and paid), an order which is not in transaction (placed and not paid) and an order which is purchased (not placed and not paid) and determining the transaction frequency of the identity information of the current purchasing user by counting the occurrence amount of the current payment identification information in the historical order information.

The preset transaction frequency threshold may include a preset current purchasing user identity information transaction frequency sub-threshold and a preset current payment identification information transaction frequency sub-threshold, and then the current purchasing user identity information transaction frequency and the current payment identification information transaction frequency are respectively compared with the preset current purchasing user identity information transaction frequency sub-threshold and the preset current payment identification information transaction frequency sub-threshold, and whether the user transaction frequency is lower than the preset transaction frequency threshold is determined according to a certain rule, for example, the sum of two values is smaller than the respective corresponding sub-threshold, or the sum of two values is smaller than the sum of two thresholds, and the like.

The current method for determining the transaction quantity of key medicines of the identity information of the purchased user comprises the following steps:

and determining the quantity of key medicines successfully transacted in the historical order information according to the identity information of the current purchase user.

The determination mode of the transaction quantity of the key medicines of the current payment identification information comprises the following steps:

and determining the quantity of key medicines successfully transacted in the historical order information according to the current payment identification information.

Wherein the key drug can be one or more, and the key drug can be selected by the skilled person according to the needs.

The preset transaction quantity threshold includes a preset current additional user identity information sub-quantity threshold and a preset current payment identification information sub-threshold, and correspondingly, whether the transaction quantity of the user key medicines is lower than the preset transaction quantity threshold is similar to whether the transaction frequency of the user is lower than the preset transaction frequency threshold, which is not repeated herein.

The transaction suspension can be long-term suspension or third-party intervention audit, and whether the transaction suspension is continued or not is determined according to the audit result. If the transaction is suspended, the drug sales list is not generated.

The continuous transaction is to continue the steps of obtaining the prescription quantity, the stock quantity and the preset limit purchase quantity of the prescription drugs in the formal prescription and generating the drug sales list.

Through the implementation mode, the waste of system resources caused by actions of purchase adding, order placing, purchase and the like of a user at a high frequency can be avoided, repeated purchase of the user is avoided by limiting multiple purchases, and the risk of sudden increase of subsequent return quantity or resale of accumulated medicines for profit is reduced.

The embodiment provides a prescription generating method based on natural language processing, which generates indication, optional use specification information, suspected symptoms and recommended use specification information by acquiring medicine purchase information and user description information, generates prescription necessary information and information credibility thereof, further generates an initial prescription, converts the language or text description information and the medicine purchase information of a user into the indication, optional use specification information, suspected symptoms and recommended use specification information by means of natural language processing technology, can effectively reduce the time for judging whether the purchased medicine is suitable for the user, can generate the initial prescription by determining the prescription necessary information and information credibility if the user clearly knows own symptoms and required medicines and determining the credibility of the information including credibility according to the data in an automatic checking manner, when the patient purchases the medicine, the experience degree and the satisfaction degree of the patient can be improved, if the credibility degree of the information comprises unreliability, necessary information of the prescription can be provided for the corresponding doctor who receives the doctor, the doctor-patient communication time is saved, the experience degree and the satisfaction degree of the patient can be improved when the patient purchases the medicine, and in addition, the doctor receiving efficiency and the medical resource utilization rate can be improved.

Doctors are the core of hospital operation, the charging of the medical system is mainly realized under the promotion of the doctors, and other processes are all used for laying the doctors or assisting the doctors to carry out work. Therefore, how the doctor's visit-maker status is critical to the hospital revenue. The final purpose of doctor's reception is to cure the disease for the patient and create economic benefit for the hospital. The doctor's efficiency of the prescription of receiving a doctor is a method for improving the satisfaction of the patient and a method for improving the medical economic benefit; or a method of balancing the relationship of the two.

In the traditional doctor consultation prescription, doctors need to face complex, tedious and tedious inquiry with patients, including but not limited to chief complaints, current medical history, past history, family history, personal history, marriage and childbearing history and the like, the doctors primarily know the economic transfer conditions, the attitude towards health, the motivation for the current diagnosis and the like of the patients, which easily causes nervous emotion to the patients, meanwhile, the doctors enter inquiry information of the patients into a system for needing time, so that a large number of patients enter a waiting period, a large amount of precious doctor resources are invested to provide simple, repeated and low-value labor, so that the work efficiency of the doctors is difficult to improve, the operation cost is high, the satisfaction of the patients is also influenced, and in order to solve the problems, the embodiment provides a prescription generation method based on natural language processing. The above-mentioned natural language processing based recipe generating method is exemplarily described below by a specific embodiment, referring to fig. 4, and the implementation flow of the specific natural language processing based recipe generating method is as follows:

a user firstly adds commodities (medicines) to a shopping cart to obtain medicine purchase information, and the medicine purchase information comprises medicines, indications, ICD10 standard code mapping relation and the like. Inquiring a user through equipment such as a robot and the like to obtain user description information, determining the chief complaint information and the user basic information (information of a patient) according to the user description information, recommending diagnosis and medicines according to the chief complaint information and the medicine purchase information, and obtaining necessary prescription information according to the diagnosis and the medicines through template context. Judging the credibility of the necessary information of the prescription, if the necessary information of the prescription is not credible, entering manual prescription when the reasons such as the administration frequency or the use amount are not appropriate, the control of mental diseases and the like do not meet the automatic prescription condition, distributing a doctor for receiving a doctor for the user, bringing the collected user description information, medicine purchase information and the necessary information of the prescription into a doctor workbench, and after the prescription authority of the doctor is verified, if the doctor has the prescription authority, manually making the initial prescription by the doctor. If the necessary information of the prescription is credible, the automatic initial prescription can be directly generated according to the necessary information of the prescription. Whether the manual initial prescription is made by a doctor, the prescription enters an auditing platform for next circulation through robot running, the manual initial prescription or the automatic initial prescription can be obtained after the process is completed, the initial prescription is audited by a third party, the formal prescription is made after the audit is passed, and the user can receive a prescription card at this moment. The user can place an order and pay to purchase the medicine after receiving the prescription card. The obtaining of the user description information and the determination of the chief complaint information can be realized by assembling the chief complaint related information in the system in advance and entering the guide. The inquiry to the user and the generation of the main complaint information can be realized by an oneshot template and the like.

By means of the method, doctor consultation operation can be simplified, the increase conversion rate of doctor prescription can be increased, doctor processing timeliness is improved, inquiry congestion is reduced, channel service capacity and channel service quantity are improved, prescription is automatically made in an AI + robot mode as far as possible, and transaction frequency is improved. If the automatic prescription is failed, the collected user basic information and the chief complaint information can be brought into a doctor workbench, so that the content required to be manually filled in when the doctor prescribes the prescription is simplified, the user appeal is quickly known, the user is better served, the service quality is improved, and the user satisfaction is ensured.

In one embodiment, the present invention further provides a natural language processing based prescription generating apparatus 500, referring to fig. 5, including:

the purchase information acquisition module 501 is configured to acquire medicine purchase information, and determine indications and optional use specification information of purchased medicines;

a description information obtaining module 502, configured to obtain user description information and determine complaint information;

a prescription necessary information generating module 503, configured to generate prescription necessary information according to the chief complaint information and the medicine purchase information, where the prescription necessary information includes at least one of suspected symptoms and recommended use specification information;

a determining module 504, configured to determine an information credibility of necessary prescription information according to at least one of symptom information and specification information, where the symptom information is determined according to an indication and a suspected symptom, and the specification information is determined according to optional use specification information and recommended use specification information;

and an initial prescription generating module 505 for generating an initial prescription according to the information credibility and the necessary prescription information.

Optionally, the determining module includes:

the symptom credibility submodule is used for determining the credibility of the symptom according to the indication and the suspected symptom;

the use specification credibility submodule is used for determining the use specification credibility according to the optional use specification information and the suggested use specification information;

the first judgment module is used for judging whether the symptom credibility and the use specification credibility both comprise credibility or not, and judging whether the information credibility comprises credibility or not;

and the second judging module is used for judging whether at least one of the symptom credibility and the use specification credibility comprises the unreliability, and judging whether the information credibility comprises the unreliability.

Optionally, the initial prescription generating module includes any one of:

the automatic generation module is used for generating an initial prescription according to the necessary information of the prescription if the credibility of the information comprises credibility;

and the manual generation module is used for acquiring user associated information if the information credibility degree comprises unreliability, determining the importance degree of the purchase adding event according to at least one of the user associated information, the user description information and the medicine purchase adding information, distributing a doctor receiving according to the importance degree of the purchase adding event, sending the necessary information of the prescription to the doctor receiving, acquiring the prescription instruction of the doctor receiving and generating the initial prescription, wherein the user associated information comprises at least one of the user position, the user purchase frequency and the historical transaction amount.

Optionally, the artificial generation module further includes:

the emotion submodule is used for determining the importance degree of the purchase adding event according to the user description information, acquiring emotion information in the user description information, determining the emotion state of the user and determining the importance degree of the purchase adding event, wherein the emotion information comprises at least one of expression, action, tone, speed and tone;

and the money duration submodule is used for determining the importance degree of the purchase event according to the medicine purchase information, and determining the importance degree of the purchase event according to the purchase amount and the waiting duration.

Optionally, the description information obtaining module includes:

the first inquiry module is used for determining symptom questions according to the indications, and inquiring for the first time to obtain a first answer result;

the keyword extraction module is used for inputting the answer result into a preset keyword extraction model to obtain initial information;

the second inquiry module is used for generating a confirmation question according to the initial information, and carrying out second inquiry and second answer to the result;

and the main complaint information generating module is used for generating the main complaint information according to the initial information if the second answer result comprises the confirmed initial information.

Optionally, the apparatus further comprises:

the checking module is used for checking the initial prescription;

the formal prescription generating module is used for generating a formal prescription according to the initial prescription if the verification passes;

and the medicine sales list generating module is used for acquiring the prescription quantity, the inventory quantity and the preset limit purchase quantity of the prescription medicines in the formal prescription and generating a medicine sales list.

The generation mode of the medicine sales list comprises any one of the following modes:

if the prescription quantity is not greater than the preset limit purchase quantity and the prescription quantity is not greater than the inventory quantity, the medicine sales list comprises prescription medicines and the prescription quantity;

if the stock quantity is greater than 0, the prescription quantity is not greater than the preset limit purchase quantity, and the stock quantity is less than the prescription quantity, the medicine sales list comprises prescription medicines and the stock quantity;

if the stock quantity is equal to 0, carrying out the shortage prompt of the prescription medicine on the medicine sales list;

and if the prescription quantity is not larger than the inventory quantity, the prescription quantity is larger than the preset limit purchase quantity, the medicine sales list comprises the prescription medicines and the preset limit purchase quantity, and the prescription quantity reduction prompt is carried out.

Optionally, the apparatus further comprises a transaction progress determination module, configured to, before the drug sales list generation module obtains the prescription quantity, the inventory quantity and the preset limit purchase quantity of the prescription drugs in the official prescription, and generates the drug sales list,

acquiring historical order information and current user identification information, wherein the current user identification information comprises at least one of current purchase adding user identity information and current payment identification information, and the historical order information comprises at least one of historical purchase adding user identity information, historical purchase adding medicines, historical purchase adding medicine quantity and historical payment identification information;

determining user transaction frequency according to historical order information and current user identification information, wherein the user transaction frequency comprises at least one of current purchase adding user identity information transaction frequency and current payment identification information transaction frequency;

if the transaction frequency of the user is higher than a preset transaction frequency threshold value, suspending the transaction;

if the user transaction frequency is lower than a preset transaction frequency threshold value, determining the user key medicine transaction quantity according to the historical order information and the current user identification information, wherein the user key medicine transaction quantity comprises at least one of the current purchase user identity information key medicine transaction quantity and the current payment identification information key medicine transaction quantity, and if the user key medicine transaction quantity exceeds the preset transaction quantity threshold value, suspending the transaction;

and if the transaction frequency of the user is lower than the preset transaction frequency threshold value and the transaction quantity of the key medicines of the user is lower than the preset transaction quantity threshold value, continuing the transaction.

The prescription generating device based on natural language processing in the embodiment generates the indication, the optional use specification information, the suspected symptom and the recommended use specification information by acquiring the medicine purchase information and the user description information, generates the necessary information and the credibility of the prescription information, and further generates the initial prescription, converts the language or text description information and the medicine purchase information of the user into the indication, the optional use specification information, the suspected symptom and the recommended use specification information by means of the natural language processing technology, can effectively reduce the judgment time of whether the purchased medicine is suitable for the user, can generate the initial prescription by determining the necessary information and the credibility of the prescription information if the user clearly knows the own symptom and the required medicine, and can generate the initial prescription in an automatic checking mode by determining the credibility of the information according to the data, when the patient purchases the medicine, the experience degree and the satisfaction degree of the patient can be improved, if the credibility degree of the information comprises unreliability, necessary information of the prescription can be provided for the corresponding doctor who receives the doctor, the doctor-patient communication time is saved, the experience degree and the satisfaction degree of the patient can be improved when the patient purchases the medicine, and in addition, the doctor receiving efficiency and the medical resource utilization rate can be improved.

It should be understood that the natural language processing based prescription generating device system is substantially provided with a plurality of modules for executing the natural language processing based prescription generating method in any of the embodiments, and specific functions and technical effects are only required to refer to the embodiments, which are not described herein again.

In an embodiment, referring to fig. 6, the embodiment further provides a computer device 600, which includes a memory 601, a processor 602, and a computer program stored on the memory and executable on the processor, and when the processor 602 executes the computer program, the steps of the method according to any one of the above embodiments are implemented.

In an embodiment, a computer-readable storage medium is also provided, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any of the above embodiments.

The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments. Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.