阅读说明：本技术 一种智能母婴知识服务方法及系统 (Intelligent mother and infant knowledge service method and system ) 是由 熊杰 金炎芳 于 2019-10-24 设计创作，主要内容包括：本发明公开了一种智能母婴知识服务方法及系统,该方法包括：周期性或实时采集大量的母婴数据信息,并储存在数据库中；所述母婴数据信息包括：母婴相关知识问题及相关知识答案；根据储存在数据库中的母婴数据信息,构建问答知识库；获取用户发送的母婴知识查询请求,将所述查询请求在所述问答知识库中查找相对应的信息；所述查询请求包括：用户问题；将查找的相对应的信息反馈给用户；本发明通过建立问答知识库,有效的解决了现有母婴知识服务方法不能完整的展现一种从孕妇备孕到孕妇分娩,再到孩子养育的全过程,缺乏人性化智能的展现方式,展现形式单一的问题。(The invention discloses an intelligent mother-infant knowledge service method and system, wherein the method comprises the following steps: collecting a large amount of mother and infant data information periodically or in real time, and storing the information in a database; the maternal and infant data information includes: mother and infant related knowledge questions and related knowledge answers; constructing a question-answer knowledge base according to mother-infant data information stored in the database; acquiring a mother-infant knowledge query request sent by a user, and searching corresponding information in the question-answer knowledge base by the query request; the query request includes: a user question; feeding back the searched corresponding information to the user; by establishing the question-answer knowledge base, the problems that the whole process from pregnancy preparation of the pregnant woman to delivery of the pregnant woman and then to child nurturing cannot be completely shown by the conventional maternal and infant knowledge service method, a humanized and intelligent showing mode is lacked, and the showing form is single are effectively solved.)

1. An intelligent mother-infant knowledge service method is characterized by comprising the following steps:

s1, collecting a large amount of mother and infant data information periodically or in real time, and storing the information in a database; the maternal and infant data information includes: mother and infant related knowledge questions and related knowledge answers;

s2, constructing a question-answer knowledge base according to the mother-infant data information stored in the database;

s3, acquiring a mother and infant knowledge query request sent by a user, and searching corresponding information in the question and answer knowledge base by the query request; the query request includes: a user question;

and S4, feeding back the searched corresponding information to the user.

2. The intelligent maternal and infant knowledge service method according to claim 1, wherein the searching the query request for corresponding information in the question-answer knowledge base comprises:

s31, preprocessing: processing the query request of the user through complex and simple conversion, segmentation of a word segmentation device, main body and field identification;

s32, search: narrowing the search domain of the question and answer knowledge base according to the main body and the domain identification result obtained in the preprocessing, calculating the semantic similarity between the user question and the question in the question and answer knowledge base, and recalling the question similar to the user question in the question and answer knowledge base to form a candidate question set;

s33, matching: measuring similarity characteristic values between the user questions and the candidate question sets according to different dimensions; the different dimensions comprise editing distance and semantic similarity;

s34, sorting: and sequencing the candidate question sets according to the similarity characteristic values, and feeding back answers of the most similar questions to the user.

3. The intelligent maternal and infant knowledge service method according to claim 2, wherein calculating semantic similarity between the user questions and the questions in the question and answer knowledge base comprises:

s321, calculating semantic vectors P for all questions in the question and answer knowledge base;

s322, constructing a corresponding semantic tree according to the generated semantic vector P, and calculating a semantic vector P' for the user problem;

s323, calculating the cosine similarity of the semantic vectors of the two problems, wherein the calculation formula is as follows:

wherein cos (PP') represents a semantic vector cosine similarity; the semantic vector cosine similarity is semantic similarity.

4. The intelligent maternal and infant knowledge service method according to claim 3, wherein the building of the corresponding semantic tree in the step S322 includes: and according to the result of the preprocessing based on the main body and the domain identification, establishing semantic tree indexes in advance for the questions in the question-answer knowledge base by using an annoy algorithm.

5. The intelligent maternal and infant knowledge service method according to claim 2, wherein the step S34 includes: and weighting the similarity characteristic values, and sequencing according to weighting results.

6. The utility model provides an intelligence mother and infant knowledge service system which characterized in that includes:

an acquisition module: the system is used for periodically or real-timely acquiring a large amount of mother and infant data information and storing the information in a database; the maternal and infant data information includes: mother and infant related knowledge questions and related knowledge answers;

constructing a module: the question and answer knowledge base is constructed according to the mother and infant data information stored in the database;

a searching module: acquiring a mother-infant knowledge query request sent by a user, and searching corresponding information in the question-answer knowledge base by the query request; the query request includes: a user question;

a feedback module: for feeding back the searched corresponding information to the user.

7. The intelligent maternal and infant knowledge service system of claim 8, wherein the lookup module comprises:

a preprocessing submodule: the query request of the user is processed through simplified and simplified conversion, segmentation of a word segmentation device, main body and field identification;

a retrieval submodule: the system comprises a question and answer knowledge base, a question and answer knowledge base and a question and answer knowledge base, wherein the question and answer knowledge base is used for preprocessing a subject and a domain identification result;

matching sub-modules: the method comprises the steps of measuring similarity characteristic values between user questions and a candidate question set according to different dimensions; the different dimensions comprise editing distance and semantic similarity;

a sorting submodule: and the candidate question sets are ranked according to the similarity characteristic values, and answers of the most similar questions are fed back to the user.

8. The intelligent maternal and infant knowledge service system of claim 7, wherein the retrieval submodule comprises:

the first calculation unit: calculating semantic vectors P for all questions in the question-answer knowledge base;

a second calculation unit: constructing a corresponding semantic tree according to the generated semantic vector P, and calculating a semantic vector P' for the user problem;

a third calculation unit: calculating the cosine similarity of the semantic vectors of the two problems, wherein the calculation formula is as follows:

wherein cos (PP') represents a semantic vector cosine similarity; the semantic vector cosine similarity is semantic similarity.

9. The intelligent maternal and infant knowledge service system according to claim 8, wherein the building of the corresponding semantic tree in the second computing unit comprises: and according to the result of the preprocessing based on the main body and the domain identification, establishing semantic tree indexes in advance for the questions in the question-answer knowledge base by using an annoy algorithm.

10. The intelligent maternal and infant knowledge service system according to claim 7, wherein the ranking submodule is configured to weight the similarity feature values and rank the similarity feature values according to a weighting result.

Technical Field

The invention belongs to the field of mother and infant health care technology and computer internet, and particularly relates to an intelligent mother and infant knowledge service method and system.

Background

With the rapid development of the internet, computer technology has been convenient for people's life in various industries. No exception is made in the medical field. A large amount of maternal and infant professional data are hidden in the network, and the information is scattered and hidden in sand like gold for solving the maternal and infant health care problem. Although useful, the system is not systematic, complete, flexible and intelligent organization and humanized display modes are lacked, and the system cannot play a good role finally.

Through investigation, the current child-care platforms such as 'child-care net', 'babytree', 'YY net', etc. in the market are more concerned about children and are more professional for child-care. However, they are only concerned with the popularization of health care and childbearing knowledge, and do not have the popularization of knowledge related to pregnant women such as preparation before pregnancy, pregnancy examination, and the like, nor have the effective construction and arrangement of information to form a knowledge base to better serve users.

Therefore, in order to make the mother-infant professional data on the network more systematic and complete, how to develop a method and a system to collect and arrange the mother-infant knowledge becomes a technical problem to be solved urgently in the industry.

Disclosure of Invention

In order to solve the problems that information cannot be effectively constructed and arranged and a user cannot be better served in the prior art, the invention provides an intelligent mother and infant knowledge service method and system.

In order to solve the above technical problem, in a first aspect, an embodiment of the present invention provides an intelligent mother and infant knowledge service method, including:

s1, collecting a large amount of mother and infant data information periodically or in real time, and storing the information in a database; the maternal and infant data information includes: mother and infant related knowledge questions and related knowledge answers;

s2, constructing a question-answer knowledge base according to the mother-infant data information stored in the database;

s3, acquiring a mother and infant knowledge query request sent by a user, and searching corresponding information in the question and answer knowledge base by the query request; the query request includes: a user question;

and S4, feeding back the searched corresponding information to the user.

In one embodiment, the searching the query request for corresponding information in the question-answer knowledge base includes:

s31, preprocessing: processing the query request of the user through complex and simple conversion, segmentation of a word segmentation device, main body and field identification;

s32, search: narrowing the search domain of the question and answer knowledge base according to the main body and the domain identification result obtained in the preprocessing, calculating the semantic similarity between the user question and the question in the question and answer knowledge base, and recalling the question similar to the user question in the question and answer knowledge base to form a candidate question set;

s33, matching: measuring similarity characteristic values between the user questions and the candidate question sets according to different dimensions; the different dimensions comprise editing distance and semantic similarity;

s34, sorting: and sequencing the candidate question sets according to the similarity characteristic values, and feeding back answers of the most similar questions to the user.

In one embodiment, calculating semantic similarity of a user question to questions in a knowledge base of questions and answers includes:

s321, calculating semantic vectors P for all questions in the question and answer knowledge base;

s322, constructing a corresponding semantic tree according to the generated semantic vector P, and calculating a semantic vector P' for the user problem;

s323, calculating the cosine similarity of the semantic vectors of the two problems, wherein the calculation formula is as follows:

wherein cos (PP') represents a semantic vector cosine similarity; the semantic vector cosine similarity is semantic similarity.

In one embodiment, the building of the corresponding semantic tree in step S322 includes: and according to the result of the preprocessing based on the main body and the domain identification, establishing semantic tree indexes in advance for the questions in the question-answer knowledge base by using an annoy algorithm.

In one embodiment, the step S34 includes: and weighting the similarity characteristic values, and sequencing according to weighting results.

In a second aspect, the present invention further provides an intelligent mother-infant knowledge service system, including:

an acquisition module: the system is used for periodically or real-timely acquiring a large amount of mother and infant data information and storing the information in a database; the maternal and infant data information includes: mother and infant related knowledge questions and related knowledge answers;

constructing a module: the question and answer knowledge base is constructed according to the mother and infant data information stored in the database;

a searching module: acquiring a mother-infant knowledge query request sent by a user, and searching corresponding information in the question-answer knowledge base by the query request; the query request includes: a user question;

a feedback module: for feeding back the searched corresponding information to the user.

In one embodiment, the lookup module comprises:

a preprocessing submodule: the query request of the user is processed through simplified and simplified conversion, segmentation of a word segmentation device, main body and field identification;

a retrieval submodule: the system comprises a question and answer knowledge base, a question and answer knowledge base and a question and answer knowledge base, wherein the question and answer knowledge base is used for preprocessing a subject and a domain identification result;

matching sub-modules: the method comprises the steps of measuring similarity characteristic values between user questions and a candidate question set according to different dimensions; the different dimensions comprise editing distance and semantic similarity;

a sorting submodule: and the candidate question sets are ranked according to the similarity characteristic values, and answers of the most similar questions are fed back to the user.

In one embodiment, the retrieval submodule includes:

the first calculation unit: calculating semantic vectors P for all questions in the question-answer knowledge base;

a second calculation unit: constructing a corresponding semantic tree according to the generated semantic vector P, and calculating a semantic vector P' for the user problem;

a third calculation unit: calculating the cosine similarity of the semantic vectors of the two problems, wherein the calculation formula is as follows:

wherein cos (PP') represents a semantic vector cosine similarity; the semantic vector cosine similarity is semantic similarity.

In one embodiment, the building of the corresponding semantic tree in the second computing unit includes: and according to the result of the preprocessing based on the main body and the domain identification, establishing semantic tree indexes in advance for the questions in the question-answer knowledge base by using an annoy algorithm.

In one embodiment, the sorting submodule is configured to weight the similarity feature values and sort according to a weighting result.

The invention has the advantages that the intelligent maternal and infant knowledge service method and system are provided, the system of maternal and infant knowledge is more complete by constructing the question-answer knowledge base, and the problems that the whole process from pregnancy preparation of a pregnant woman to delivery of the pregnant woman and then to child nurturing cannot be completely shown by the existing maternal and infant knowledge service method, a humanized and intelligent showing mode is lacked, and the showing form is single are solved.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

fig. 1 is a flowchart of an intelligent mother-infant knowledge service method provided by the embodiment of the invention.

Fig. 2 is a flowchart of searching for corresponding information according to an embodiment of the present invention.

Fig. 3 is a flowchart for calculating semantic similarity according to an embodiment of the present invention.

Fig. 4 is a diagram of an intelligent mother-infant knowledge service system provided by the invention.

Fig. 5 is a flowchart illustrating the operation of the search module in the intelligent maternal and infant knowledge service system provided by the present invention.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

The invention provides an intelligent mother-infant knowledge service method, which is shown in figure 1 and comprises the following steps:

s1, collecting a large amount of mother and infant data information periodically or in real time, and storing the information in a database; the mother-infant data information includes: mother and infant related knowledge questions and related knowledge answers;

s2, constructing a question-answer knowledge base according to the mother-infant data information stored in the database;

s3, acquiring a mother and infant knowledge query request sent by a user, and searching corresponding information in a question and answer knowledge base by the query request; the query request includes: a user question;

and S4, feeding back the searched corresponding information to the user.

In the present embodiment, three parties are involved: the system comprises a user client, a system server and a database server connected with the system server. Wherein: the user client may be a PC or a mobile terminal, for example.

Information gathering in step S1 is the first pass that the system server runs in the background, and although not directly visible to the user, it provides material to the knowledge building module. Therefore, the importance of the server as a whole system is self-evident. The information can be updated in real time through periodic or real-time information acquisition to obtain the latest data information in the field, and the information on the website can be acquired, and books related to the mother and infant knowledge can be acquired. For example, information on a website can be collected and downloaded at regular time by using a crawler program, and latest maternal and infant data can be analyzed and stored in a database.

The information acquisition process comprises the following steps: for example, firstly, selecting a plurality of relatively authoritative websites (mom web, pacific parent web, nursery web and the like), then crawling the data regularly, then analyzing, removing useless tags and advertisements, and finally persisting in a database; the crawler program is a program for automatically downloading web pages, and selectively accesses the web pages and related links on the world wide web to acquire required information according to a set grabbing target. Rather than pursuing large coverage, the crawler targets the crawl of web pages related to content on a particular topic, preparing data resources for topic-oriented user queries. The embodiment of the invention does not limit the source of the collection.

In the steps S2-S4, a question-answer knowledge base is constructed according to mother-infant data information stored in the database, so that the related data information of the mother and the infant is more complete, after user questions are obtained, the user questions are preprocessed, the processed results are retrieved in the question-answer knowledge base, feature matching is conducted on the questions retrieved in the question-answer knowledge base and the questions of the user, sorting is conducted according to the matched results, and the question answers most similar to the user questions are fed back to the user.

As shown in fig. 2, in one embodiment, the searching the query request for corresponding information in the question-answering knowledge base includes:

s31, preprocessing: processing the query request of the user through complex and simple conversion, segmentation of a word segmentation device, main body and field identification;

s32, search: narrowing the search domain of the question and answer knowledge base according to the main body and the domain identification result obtained in the preprocessing, calculating the semantic similarity between the user question and the question in the question and answer knowledge base, and recalling the question similar to the user question in the question and answer knowledge base to form a candidate question set;

s33, matching: measuring similarity characteristic values between the user questions and the candidate question sets according to different dimensions; the different dimensions include edit distance and semantic similarity.

S34, sorting: and sequencing the candidate question sets according to the similarity characteristic values, and feeding back answers of the most similar questions to the user.

In this embodiment, in step S31, operations such as complex and simple conversion are performed on the user question, word segmentation is performed on the user question, and a subject and field recognition result of the user question is finally obtained according to the subject and field recognition. The word segmentation principle of the word segmentation device is as follows: the method is characterized in that a character string to be analyzed is matched with a vocabulary entry in a 'sufficiently large' machine dictionary according to a certain strategy, if a certain character string is found in the dictionary, the matching is successful (a word is recognized), a main body and a field in a user problem can be more easily recognized through word segmentation processing, and the search range is narrowed.

In step S32, the search range of questions in the question-and-answer knowledge base is narrowed down by the subject of the user question and the domain recognition result, for example, if the subject of the question is a pregnant woman and the domain recognition result is a pregnancy check, then the search is performed only under the corresponding question set. The retrieval mode is to calculate the semantic similarity between the user questions and the question sets, arrange the user questions and the question sets from large to small according to the semantic similarity, take the question sets corresponding to the front as candidate question sets, and if the former one hundred corresponding questions are selected, form the candidate question sets by the questions.

In steps S33-S34, similarity feature values between the user question and the candidate question set are measured through two dimensions of editing distance and semantic similarity, the similarity feature values are weighted and then are ranked, and the answer of the most similar question is returned to the user.

The editing distance refers to the minimum number of editing operations required for converting one character string into another character string. Permitted editing operations include replacing one character with another, inserting one character, and deleting one character. Generally, the smaller the edit distance, the greater the similarity of two character strings.

In one embodiment, as shown in FIG. 3, calculating semantic similarity of user questions to questions in the question-and-answer knowledge base includes:

s321, calculating semantic vectors P for all questions in a question and answer knowledge base;

s322, constructing a corresponding semantic tree according to the generated semantic vector P, and calculating a semantic vector P' for the user problem;

s323, calculating the cosine similarity of the semantic vectors of the two problems, wherein the calculation formula is as follows:

wherein cos (PP') represents a semantic vector cosine similarity; the semantic vector cosine similarity is the semantic similarity.

The building of the corresponding semantic tree in step S322 specifically includes: and according to the result of the preprocessing based on the main body and the domain identification, establishing semantic tree indexes in advance for the questions in the question-answer knowledge base by using an annoy algorithm.

The step S34 includes: and weighting the similarity characteristic values, and sequencing according to weighting results.

According to the intelligent maternal-infant knowledge service method disclosed by the invention, by establishing the question-answer knowledge base, the problems that the whole process from the pregnancy preparation of the pregnant woman to the delivery of the pregnant woman and then to the nurturing of the child cannot be completely shown by the existing maternal-infant knowledge service method, a humanized and intelligent showing mode is lacked, and the showing form is single are effectively solved. According to the method, a large amount of maternal and infant professional data are effectively constructed and sorted, so that the information related to the maternal and infant is more complete, the matching accuracy is improved by calculating the semantic similarity between the user questions and the questions in the question and answer knowledge base, the similarity characteristic values are weighted, the ranking is carried out according to the weighting results, the ranking is finally fed back to the user, and the use experience of the user is improved.

Based on the same inventive concept, the embodiment of the invention also provides an intelligent mother and infant knowledge service system, and as the principle of the problem solved by the system is similar to that of the method, the implementation of the system can refer to the implementation of the method, and repeated details are not repeated.

In a second aspect, the present invention further provides a system for providing an intelligent maternal and infant knowledge service, which is shown in fig. 4, and includes:

the acquisition module 1: the system is used for periodically or real-timely acquiring a large amount of mother and infant data information and storing the information in a database; the maternal and infant data information includes: mother and infant related knowledge questions and related knowledge answers;

and constructing a module 2: the question and answer knowledge base is constructed according to the mother and infant data information stored in the database;

the searching module 3: acquiring a mother-infant knowledge query request sent by a user, and searching corresponding information in a question-answer knowledge base by the query request; the query request includes: a user question;

the feedback module 4: for feeding back the searched corresponding information to the user.

As shown in fig. 4-5, in one embodiment, the lookup module 3 includes:

the preprocessing submodule 31: the query processing system is used for processing a query request of a user through simplified and simplified conversion, segmentation of a word segmentation device, main body and field identification;

the retrieval submodule 32: the system comprises a question and answer knowledge base, a question and answer knowledge base and a question and answer knowledge base, wherein the question and answer knowledge base is used for preprocessing a subject and a domain identification result;

matching sub-module 33: the method comprises the steps of measuring similarity characteristic values between user questions and a candidate question set according to different dimensions; different dimensions include edit distance and semantic similarity;

the ordering sub-module 34: and the method is used for sequencing the candidate question sets according to the similarity characteristic values and feeding back answers of the most similar questions to the user.

In one embodiment, the retrieval submodule 32 includes:

the first calculation unit 321: calculating semantic vectors P for all questions in a question-and-answer knowledge base;

the second calculation unit 322: constructing a corresponding semantic tree according to the generated semantic vector P, and calculating a semantic vector P' for the user problem;

the third calculation unit 323: calculating the cosine similarity of the semantic vectors of the two problems, wherein the calculation formula is as follows:

wherein cos (PP') represents a semantic vector cosine similarity; the semantic vector cosine similarity is the semantic similarity.

In one embodiment, the second computing unit 322 constructs a corresponding semantic tree, including: and according to the result of the preprocessing based on the main body and the domain identification, establishing semantic tree indexes in advance for the questions in the question-answer knowledge base by using an annoy algorithm.

In one embodiment, the sorting sub-module 34 is configured to weight the similarity characteristic values and sort according to the weighted result.

According to the intelligent maternal and infant knowledge service system provided by the invention, maternal and infant information is effectively constructed and arranged and presented to a user in a question-answer knowledge base mode, and the problems that the whole process from pregnancy preparation of a pregnant woman to delivery of the pregnant woman and then to child nurturing cannot be completely presented by the existing maternal and infant knowledge service method, a humanized intelligent presentation mode is lacked, and the presentation mode is single are solved.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.