阅读说明：本技术 一种基于ner和nlu的骚扰信息判断方法及系统 (Disturbance information judgment method and system based on NER and NLU ) 是由 张超 于 2020-12-01 设计创作，主要内容包括：本发明公开了一种基于NER和NLU的骚扰信息判断方法及系统,其中,所述方法包括：获得第一文本信息；对所述第一文本信息进行自然语言理解处理,获得自然语言理解意图列表信息；对所述第一文本信息进行命名实体识别,获得词槽内容信息；根据所述自然语言理解意图列表信息和\或所述词槽内容信息,获得第一结果信息；判断所述第一结果信息是否包含骚扰信息；如果所述第一结果信息包含骚扰信息,获得第一标记信息；根据所述第一标记信息,将所述第一文本信息标记为骚扰信息。解决了现有技术极大的耗费时间与精力,无法做到迅速响应,针对性不强,无法做到低成本更新的技术问题。(The invention discloses a disturbance information judgment method and a system based on NER and NLU, wherein the method comprises the following steps: obtaining first text information; performing natural language understanding processing on the first text information to obtain natural language understanding intention list information; carrying out named entity recognition on the first text information to obtain word slot content information; acquiring first result information according to the natural language understanding intention list information and/or the word slot content information; judging whether the first result information contains harassment information or not; if the first result information contains harassment information, first mark information is obtained; and marking the first text information as harassment information according to the first marking information. The technical problems that in the prior art, time and energy are greatly consumed, rapid response cannot be achieved, pertinence is not strong, and low-cost updating cannot be achieved are solved.)

1. A disturbance information judgment method based on NER and NLU is disclosed, wherein the method comprises the following steps:

obtaining first text information;

performing natural language understanding processing on the first text information to obtain natural language understanding intention list information;

carrying out named entity recognition on the first text information to obtain word slot content information;

acquiring first result information according to the natural language understanding intention list information and/or the word slot content information;

judging whether the first result information contains harassment information or not;

if the first result information contains harassment information, first mark information is obtained;

and marking the first text information as harassment information according to the first marking information.

2. The method of claim 1, wherein the obtaining first textual information comprises:

obtaining first information;

judging format information of the first information;

if the first information is first text information, natural language understanding processing is carried out on the first text information;

if the first information is audio information, obtaining first conversion information;

and converting the audio information into first text information according to the first conversion information.

3. The method of claim 1, wherein the performing natural language understanding processing on the first text information to obtain natural language understanding intention list information comprises:

obtaining a harassment corpus training set;

obtaining log data;

inputting the data in the disturbance corpus training set and the log data as input data into a disturbance model for training to obtain output information;

and acquiring the natural language understanding intention list information according to the output information.

4. The method of claim 1, wherein the performing named entity recognition on the first text information to obtain word slot content information comprises:

obtaining a regular entity according to the harassment word rule;

obtaining a dictionary entity according to the harassment dictionary library;

taking the harassment dictionary base and the training data set as training data to obtain a model entity;

performing word slot filling on the regular entity and the dictionary entity through matching logic to obtain first word slot content information;

filling word slots in the model entity through model prediction to obtain second word slot content information;

and obtaining the word slot content information according to the first word slot content information and the second word slot content information.

5. The method as claimed in claim 1, wherein the obtaining of the first result information according to the natural language understanding intention list information and/or the word slot content information comprises:

judging whether the natural language understanding intention list information and the word slot content information both have calculation results;

if only the natural language understanding intention list information or the word slot content information has a calculation result, the party having the calculation result is obtained as first result information.

6. The method of claim 5, wherein said determining whether the natural language understanding intention list information and the word slot content information both have a calculation result comprises:

if the natural language understanding intention list information and the word slot content information both have calculation results, judging whether an intention of which the degree of approximation is greater than a predetermined threshold exists in the natural language understanding intention list information;

if there is an intention in the natural language understanding intention list information whose degree of approximation is greater than a predetermined threshold value, a calculation result of the natural language understanding intention list information is taken as first result information.

7. The method according to claim 6, wherein said determining whether there is an intention in the natural language understanding intention list information whose degree of approximation is greater than a predetermined threshold value comprises:

and if no intention with the approximation degree larger than a preset threshold exists in the natural language understanding intention list information, taking the calculation result of the word slot content information as first result information.

8. A disturbance information judgment system based on NER and NLU, wherein the system comprises:

a first obtaining unit configured to obtain first text information;

a second obtaining unit configured to perform natural language understanding processing on the first text information to obtain natural language understanding intention list information;

a third obtaining unit, configured to perform named entity identification on the first text information to obtain word slot content information;

a fourth obtaining unit, configured to obtain first result information according to the natural language understanding intention list information and/or the word slot content information;

the first judging unit is used for judging whether the first result information contains harassment information;

a fifth obtaining unit, configured to obtain first marker information if the first result information includes disturbance information;

the first marking unit is used for marking the first text information as harassment information according to the first marking information.

9. A system for determining disturbance information based on NER and NLU, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 7 when executing the program.

Technical Field

The invention relates to the field of information identification, in particular to a disturbance information judgment method and system based on NER and NLU.

Background

With the rapid development of the information industry, producers of harassing information are not only humans, but also robots, which means that higher requirements are put on harassing information identification schemes, including faster response speed, more accurate identification rate, smaller errors, stronger compatibility characteristics, lower-cost lexicon updating capability and the like.

However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the inventors of the present application find that the above-mentioned technology has at least the following technical problems:

the prior art is time and energy consuming, cannot respond quickly, is not strong in pertinence and cannot update at low cost.

Disclosure of Invention

The embodiment of the application provides a disturbance information judgment method and system based on NER and NLU, solves the technical problems that in the prior art, time and energy are greatly consumed, rapid response cannot be achieved, pertinence is not strong, low-cost updating cannot be achieved, and the technical effects of higher response speed, more accurate recognition rate, smaller errors and lower-cost compatible characteristics of information recognition are achieved.

In view of the foregoing problems, embodiments of the present application provide a method and a system for determining disturbance information based on an NER and an NLU.

In a first aspect, an embodiment of the present application provides a disturbance information determination method based on an NER and an NLU, where the method includes: obtaining first text information; performing natural language understanding processing on the first text information to obtain natural language understanding intention list information; carrying out named entity recognition on the first text information to obtain word slot content information; acquiring first result information according to the natural language understanding intention list information and/or the word slot content information; judging whether the first result information contains harassment information or not; if the first result information contains harassment information, first mark information is obtained; and marking the first text information as harassment information according to the first marking information.

On the other hand, the application also provides a disturbance information judgment system based on the NER and the NLU, and the system comprises: a first obtaining unit configured to obtain first text information; a second obtaining unit configured to perform natural language understanding processing on the first text information to obtain natural language understanding intention list information; a third obtaining unit, configured to perform named entity identification on the first text information to obtain word slot content information; a fourth obtaining unit, configured to obtain first result information according to the natural language understanding intention list information and/or the word slot content information; the first judging unit is used for judging whether the first result information contains harassment information; a fifth obtaining unit, configured to obtain first marker information if the first result information includes disturbance information; the first marking unit is used for marking the first text information as harassment information according to the first marking information.

In a third aspect, the present invention provides a system for determining disturbance information based on NER and NLU, including a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method according to the first aspect when executing the program.

One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

because the natural language understanding processing is carried out on the first text information, the natural language understanding intention list information is obtained, and the named entity recognition is carried out on the first text information, so that the word slot content information is obtained. According to the natural language understanding intention list information and/or the word slot content information, first result information is obtained, if the first result information contains harassment information, first mark information is obtained, and according to the first mark information, the first text information is marked as harassment information, so that the technical effects of higher response speed, more accurate recognition rate, smaller error and lower cost of information recognition are achieved.

The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.

Drawings

Fig. 1 is a schematic flow chart of a disturbance information determination method based on NER and NLU in an embodiment of the present application;

fig. 2 is a schematic structural diagram of a disturbance information determination system based on the NER and the NLU according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of an exemplary electronic device according to an embodiment of the present application.

Description of reference numerals: a first obtaining unit 11, a second obtaining unit 12, a third obtaining unit 13, a fourth obtaining unit 14, a first judging unit 15, a fifth obtaining unit 16, a first marking unit 17, a bus 300, a receiver 301, a processor 302, a transmitter 303, a memory 304, and a bus interface 306.

Detailed Description

The embodiment of the application provides a disturbance information judgment method and system based on NER and NLU, solves the technical problems that in the prior art, time and energy are greatly consumed, rapid response cannot be achieved, pertinence is not strong, low-cost updating cannot be achieved, and the technical effects of higher response speed, more accurate recognition rate, smaller errors and lower-cost compatible characteristics of information recognition are achieved. Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are merely some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments described herein.

Summary of the application

With the rapid development of the information industry, producers of harassing information are not only humans, but also robots, which means that higher requirements are put on harassing information identification schemes, including faster response speed, more accurate identification rate, smaller errors, stronger compatibility characteristics, lower-cost lexicon updating capability and the like. However, the prior art is time and energy consuming, cannot respond quickly, is not highly targeted, and cannot update at low cost.

In view of the above technical problems, the technical solution provided by the present application has the following general idea:

the embodiment of the application provides a disturbance information judgment method based on NER and NLU, which comprises the following steps: obtaining first text information; performing natural language understanding processing on the first text information to obtain natural language understanding intention list information; carrying out named entity recognition on the first text information to obtain word slot content information; acquiring first result information according to the natural language understanding intention list information and/or the word slot content information; judging whether the first result information contains harassment information or not; if the first result information contains harassment information, first mark information is obtained; and marking the first text information as harassment information according to the first marking information.

Having thus described the general principles of the present application, various non-limiting embodiments thereof will now be described in detail with reference to the accompanying drawings.

Example one

As shown in fig. 1, an embodiment of the present application provides a disturbance information determination method based on an NER and an NLU, where the method includes:

step S100: obtaining first text information;

specifically, the first text information is source text information which needs to be judged, and includes information, mails, instant chat software (IM) and the like, and if the obtained source information is a voice type, the source text information is converted into text information through an audio identification text-to-text technology.

Step S200: performing natural language understanding processing on the first text information to obtain natural language understanding intention list information;

specifically, the Natural Language Understanding (NLU) refers to a mapping from a Natural Language to the inside of a machine, and refers to a process in which the machine can perform some Language functions desired by a human being, and the Natural Language Understanding process is to acquire an Understanding of the first text information through Natural Language Understanding. The natural language understanding intention list information is obtained by dividing the first text information into corresponding intention category lists in a classification mode after natural language understanding processing, and judging which field the first text information belongs to, wherein different intentions can have different field dictionaries, such as book names, song names, trade names and the like.

Step S300: carrying out named entity recognition on the first text information to obtain word slot content information;

specifically, the Named Entity Recognition (NER) is to recognize an Entity having a specific meaning in the first text information, and mainly includes a person name, a place name, a mechanism name, a proper noun, and the like, and mainly obtains understanding of some specific words in a sentence. The word slot content information is some necessary limiting conditions which need to be understood by the system in the user dialog and affects the execution result of the dialog system, such as: the "central station" in the "zapping" is a "tv channel word slot" which will to some extent influence the system's execution of the "zapping" intention.

Step S400: acquiring first result information according to the natural language understanding intention list information and/or the word slot content information;

specifically, the first result information is obtained by performing processing calculation of natural language understanding and named entity recognition on the first text information at the same time, and the first result information is used as respective scoring basis for judging whether harassment information is included.

Step S500: judging whether the first result information contains harassment information or not;

specifically, calculation results of natural language understanding and named entity recognition are put into a scorer to be subjected to score calculation and judgment, and whether the first result information contains harassing information or not is judged, so that the technical problems that in the prior art, time and energy are greatly consumed, quick response cannot be achieved, pertinence is not strong, and low-cost updating cannot be achieved are solved.

Step S600: if the first result information contains harassment information, first mark information is obtained;

specifically, the first marking information is used for marking disturbance information contained in the first result information text obtained through calculation, and then subsequent information identification processing is completed, so that the technical effects of faster response speed, more accurate identification rate, smaller error and lower cost of information identification are achieved.

Step S700: according to the first marking information, marking the first text information as harassment information;

specifically, if the first result information contains disturbance information, the first text information is marked as disturbance information according to the first marking information, otherwise, the process is skipped. Therefore, the technical problems that time and energy are greatly consumed, rapid response cannot be achieved, pertinence is not strong, and low-cost updating cannot be achieved in the prior art are solved, and the technical effects of higher response speed, more accurate recognition rate, smaller error and lower-cost compatible characteristics of information recognition are achieved.

Further, in the obtaining of the first text information, an embodiment of the present application further includes:

step S810: obtaining first information;

step S820: judging format information of the first information;

step S830: if the first information is first text information, natural language understanding processing is carried out on the first text information;

step S840: if the first information is audio information, obtaining first conversion information;

step S850: and converting the audio information into first text information according to the first conversion information.

Specifically, the first information is source information that needs to be determined, the format information of the first information is a source format of the first information, and two situations may exist in an information source format: voice classes including telephone, voice chat, etc., and text classes including messages, mail, instant chat software (IM), etc. And if the first information is first text information, performing next natural language understanding processing, if the first information is audio information, obtaining first conversion information, wherein the first conversion information is text information converted from an audio file by an audio identification and character conversion technology, and converting the audio information into the first text information according to the first conversion information.

Further, in the embodiment of the present application, the performing natural language understanding processing on the first text information to obtain natural language understanding intention list information further includes:

step S910: obtaining a harassment corpus training set;

step S920: obtaining log data;

step S930: inputting the data in the disturbance corpus training set and the log data as input data into a disturbance model for training to obtain output information;

step S940: obtaining the natural language understanding intention list information according to the output information;

specifically, the harassment corpus training set is a training corpus of harassment information and comprises corpus grabbing, corpus clearing, corpus preprocessing and the like, and the log data are process event record data generated by a recording system, are main training sources of the model and are beneficial to self-learning and precision improvement of the model.

Further, the disturbance model is a Neural network model, i.e., a Neural network model in machine learning, and a Neural Network (NN) is a complex Neural network system formed by widely connecting a large number of simple processing units (called neurons), which reflects many basic features of human brain functions, and is a highly complex nonlinear dynamical learning system. Neural network models are described based on mathematical models of neurons. Artificial Neural Networks (Artificial Neural Networks) are a description of the first-order properties of the human brain system. Briefly, it is a mathematical model. And inputting the data in the harassing corpus training set and the log data into a neural network model through training of a large amount of training data, and outputting the natural language understanding intention list information meeting the data in the harassing corpus training set and the log data.

Furthermore, the training process is essentially a supervised learning process, each group of supervised data comprises data in the harassing corpus training set, log data and identification information for identifying the data meeting the harassing corpus training set and the natural language understanding intention list information of the log data, the data in the harassing corpus training set and the log data are input into a neural network model, the neural network model is continuously corrected and adjusted by self according to the identification information for identifying the data meeting the harassing corpus training set and the natural language understanding intention list information of the log data until the obtained output information is consistent with the identification information, the group of supervised learning is finished, and the next group of supervised learning is carried out; and when the output information of the neural network model reaches the preset accuracy rate/reaches the convergence state, finishing the supervised learning process. Through right the supervision study of neural network model, and then make the neural network model is handled input information is more accurate, and then solves the very big time and energy that expends of prior art, can't accomplish quick response, and the pertinence is not strong, can't accomplish the technical problem of low-cost renewal, reaches the response speed that is faster to information identification, more accurate discernment rate and littleer error, the technical effect of the compatible characteristic of lower cost.

Further, in the embodiment of the present application, the named entity recognition is performed on the first text information to obtain word slot content information, and the method further includes:

step S1010: obtaining a regular entity according to the harassment word rule;

step S1020: obtaining a dictionary entity according to the harassment dictionary library;

step S1030: taking the harassment dictionary base and the training data set as training data to obtain a model entity;

step S1040: performing word slot filling on the regular entity and the dictionary entity through matching logic to obtain first word slot content information;

step S1050: filling word slots in the model entity through model prediction to obtain second word slot content information;

step S1060: and obtaining the word slot content information according to the first word slot content information and the second word slot content information.

Specifically, the harassing word rules are a system depending on manual rules, each harassing word rule is subjected to weight assistance by combining a named entity library, and then type judgment is performed according to the condition that the entities conform to the rules. The regular entities are regular expression pattern extraction entities based on the provided regular expression patterns, rule configuration from disturbance words is achieved, and the rules are obtained according to the disturbance word rules. The harassment dictionary base is an existing harassment word recognition database, and the dictionary entity is from the harassment dictionary base and is entity structure information extracted from the harassment dictionary, and the entity structure information comprises multiple variants of the same entity and the same name of different entities. The training data set is a data set used for training a data mining model in the data mining process, the disturbance dictionary base and the training data set are used as training data, the training data are respectively from the disturbance dictionary base and the training data set, the model entity is obtained, the model entity is a neural network model and is a mathematical model, geometric information and topological information can be completely represented, various operations such as Euler operation, physical property calculation, finite element analysis and the like can be supported, and the information content is complete and comprehensive. The three entities are finally calculated in two ways, the regular entity and the dictionary entity perform word slot filling through matching logic, the word slot filling is namely a sequence labeling problem, namely, corresponding labels are respectively marked on each word in a given sentence, so that necessary limiting conditions which need to be understood by a system in user conversation are obtained, the execution result of a conversation system is influenced, namely the execution result is the content information of the first word slot, the model entity fills the word slot through a model prediction function, so that the content information of the second word slot is obtained, and the content information of the word slot is obtained through comprehensive analysis according to the content information of the first word slot and the content information of the second word slot.

Further, in an embodiment of the present application, the obtaining first result information according to the natural language understanding intention list information and/or the word slot content information further includes:

step S1110: judging whether the natural language understanding intention list information and the word slot content information both have calculation results;

step S1120: if only the natural language understanding intention list information or the word slot content information has a calculation result, the party having the calculation result is obtained as first result information.

Specifically, it is determined whether there is a calculation result for both natural language understanding and named entity recognition, and if one of the natural language understanding intention list information and the word slot content information has a calculation result only for the natural language understanding intention list information or the word slot content information, only a hit result is output as first result information.

Further, after determining whether both the natural language understanding intention list information and the word slot content information have the calculation result, step S1120 in this embodiment of the present application further includes:

step S1121: if the natural language understanding intention list information and the word slot content information both have calculation results, judging whether an intention of which the degree of approximation is greater than a predetermined threshold exists in the natural language understanding intention list information;

step S1122: if there is an intention in the natural language understanding intention list information whose degree of approximation is greater than a predetermined threshold value, a calculation result of the natural language understanding intention list information is taken as first result information.

Specifically, if both natural language understanding and named entity recognition are hit, and both the natural language understanding intention list information and the word slot content information have calculation results, it is determined whether there is an intention with an approximation degree greater than 90% in the natural language understanding calculated intention list, and if there is an intention with an approximation degree greater than 90% in the natural language understanding intention list information, the recognition result of natural language understanding is selected as the first result information.

Further, after determining whether there is an intention in the natural language understanding intention list information, where the degree of approximation is greater than a predetermined threshold, step S1122 in this embodiment of the present application further includes:

step S11221: and if no intention with the approximation degree larger than a preset threshold exists in the natural language understanding intention list information, taking the calculation result of the word slot content information as first result information.

Specifically, if both the natural language understanding and the named entity recognition are hit and both have calculation results, it is determined whether there is an intention with an approximation degree greater than 90% in the list of intentions calculated by the natural language understanding, and if there is no intention with an approximation degree greater than 90% by the natural language understanding, the recognition result of the named entity recognition is preferentially adopted as the first result information.

To sum up, the disturbance information determination method and system based on the NER and the NLU provided by the embodiment of the present application have the following technical effects:

1. because the natural language understanding processing is carried out on the first text information, the natural language understanding intention list information is obtained, and the named entity recognition is carried out on the first text information, so that the word slot content information is obtained. According to the natural language understanding intention list information and/or the word slot content information, first result information is obtained, if the first result information contains harassment information, first mark information is obtained, and according to the first mark information, the first text information is marked as harassment information, so that the technical effects of higher response speed, more accurate recognition rate, smaller error and lower cost of information recognition are achieved.

2. Due to the adoption of the recognition architecture combining the two schemes of Natural Language Understanding (NLU) and entity naming recognition (NER), the two schemes are selected and combined to fully utilize the characteristics of the two schemes for complementation from the functional point of view, so that the technical effects of faster response speed, more accurate recognition rate, smaller error and lower cost of compatible characteristics of information recognition are achieved.

Example two

Based on the same inventive concept as the disturbance information judgment method based on the NER and the NLU in the foregoing embodiment, the present invention further provides a disturbance information judgment system based on the NER and the NLU, as shown in FIG. 2, the system includes:

a first obtaining unit 11, where the first obtaining unit 11 is configured to obtain first text information;

a second obtaining unit 12, where the second obtaining unit 12 is configured to perform natural language understanding processing on the first text information to obtain natural language understanding intention list information;

a third obtaining unit 13, where the third obtaining unit 13 is configured to perform named entity identification on the first text information to obtain word slot content information;

a fourth obtaining unit 14, where the fourth obtaining unit 14 is configured to obtain first result information according to the natural language understanding intention list information and/or the word slot content information;

a first judging unit 15, where the first judging unit 15 is configured to judge whether the first result information includes disturbance information;

a fifth obtaining unit 16, where the fifth obtaining unit 16 is configured to obtain first mark information if the first result information includes disturbance information;

the first marking unit 17 is configured to mark the first text information as harassment information according to the first marking information.

Further, the system further comprises:

a sixth obtaining unit configured to obtain the first information;

a second judgment unit configured to judge format information of the first information;

a first processing unit, configured to perform natural language understanding processing on the first text information if the first information is the first text information;

a seventh obtaining unit configured to obtain first conversion information if the first information is audio information;

a first conversion unit configured to convert the audio information into first text information according to the first conversion information.

Further, the system further comprises:

an eighth obtaining unit, configured to obtain a harassment corpus training set;

a ninth obtaining unit configured to obtain log data;

a tenth obtaining unit, configured to input data in the harassment corpus training set and the log data as input data into a harassment model for training, and obtain output information;

an eleventh obtaining unit configured to obtain the natural language understanding intention list information according to the output information.

Further, the system further comprises:

a twelfth obtaining unit, configured to obtain a regular entity according to the harassment word rule;

a thirteenth obtaining unit, configured to obtain a dictionary entity according to the harassment dictionary library;

a fourteenth obtaining unit, configured to obtain a model entity by using the disturbance dictionary database and the training data set as training data;

a fifteenth obtaining unit, configured to perform word slot filling on the regular entity and the dictionary entity through matching logic to obtain first word slot content information;

a sixteenth obtaining unit, configured to perform word slot filling on the model entity through model prediction to obtain second word slot content information;

a seventeenth obtaining unit, configured to obtain the word slot content information according to the first word slot content information and the second word slot content information.

Further, the system further comprises:

a third determination unit configured to determine whether the natural language understanding intention list information and the word slot content information both have a calculation result;

an eighteenth obtaining unit configured to obtain, if only the natural language understanding intention list information or the word slot content information has a calculation result, the party having the calculation result as first result information.

Further, the system further comprises:

a fourth judgment unit configured to judge whether there is an intention whose degree of approximation is larger than a predetermined threshold in the natural language understanding intention list information if both the natural language understanding intention list information and the word slot content information have a calculation result;

a first result unit configured to take a calculation result of the natural language understanding intention list information as first result information if there is an intention in the natural language understanding intention list information whose degree of approximation is greater than a predetermined threshold.

Further, the system further comprises:

a second result unit for taking a calculation result of the word bin content information as first result information if there is no intention in the natural language understanding intention list information whose degree of approximation is greater than a predetermined threshold.

Various changes and specific examples of the disturbance information judgment method based on the NER and the NLU in the first embodiment of fig. 1 are also applicable to the disturbance information judgment system based on the NER and the NLU in the present embodiment, and through the foregoing detailed description of the disturbance information judgment method based on the NER and the NLU, those skilled in the art can clearly know the implementation method of the disturbance information judgment system based on the NER and the NLU in the present embodiment, so for the brevity of the description, detailed description is not repeated here.

Exemplary electronic device

The electronic device of the embodiment of the present application is described below with reference to fig. 3.

Fig. 3 illustrates a schematic structural diagram of an electronic device according to an embodiment of the present application.

Based on the inventive concept of the disturbance information judgment method based on the NER and the NLU in the previous embodiment, the invention also provides a disturbance information judgment system based on the NER and the NLU, wherein a computer program is stored on the disturbance information judgment system, and when the program is executed by a processor, the steps of any one of the above disturbance information judgment methods based on the NER and the NLU are realized.

Where in fig. 3 a bus architecture (represented by bus 300), bus 300 may include any number of interconnected buses and bridges, bus 300 linking together various circuits including one or more processors, represented by processor 302, and memory, represented by memory 304. The bus 300 may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface 306 provides an interface between the bus 300 and the receiver 301 and transmitter 303. The receiver 301 and the transmitter 303 may be the same element, i.e., a transceiver, providing a means for communicating with various other systems over a transmission medium.

The processor 302 is responsible for managing the bus 300 and general processing, and the memory 304 may be used for storing data used by the processor 302 in performing operations.

The embodiment of the invention provides a disturbance information judgment method based on NER and NLU, which comprises the following steps: obtaining first text information; performing natural language understanding processing on the first text information to obtain natural language understanding intention list information; carrying out named entity recognition on the first text information to obtain word slot content information; acquiring first result information according to the natural language understanding intention list information and/or the word slot content information; judging whether the first result information contains harassment information or not; if the first result information contains harassment information, first mark information is obtained; and marking the first text information as harassment information according to the first marking information. The technical problems that in the prior art, time and energy are greatly consumed, rapid response cannot be achieved, pertinence is not strong, low-cost updating cannot be achieved are solved, and the technical effects of high response speed, accurate recognition rate, small errors and low-cost compatibility of information recognition are achieved.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a system for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction system which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.

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.