News event classification method

Abstract

The invention discloses a news event classification method. The method comprises the steps of obtaining text data of news events; extracting a plurality of keywords in the text data by utilizing a keyword extraction fusion model to obtain keyword phrases; performing multi-stage similarity calculation on the keyword phrases on the basis of a multi-level keyword library to obtain a similarity calculation result; performing normalization processing on the similarity calculation result, and obtaining a probability distribution result of the news events based on a multilayer sensing network model; and based on the probability distribution result, comparing a difference value of the probability values of the two news events with the highest probability value and the second highest probability value with a preset probability threshold value to obtain a classification result of the news events. According to the method, a more accurate, comprehensive and explainable news event classification result can be provided by comprehensively utilizing different types of data, a fusion model, a multi-level word bank, similarity calculation, normalization processing and flexible classification result processing.

Description

Technical Field

[0001] The present invention relates to the field of natural language processing, and more specifically to a news event classification method. Background Art

[0002] By classifying news events, readers can more easily find content they are interested in and save reading time. It can help news organizations or platforms better organize and manage large amounts of news content, making it easier to retrieve and archive. Classification can help monitor and analyze various events, and provide decision-making support and public opinion feedback for governments, businesses, and institutions.

[0003] Currently, traditional methods may only use the original text of news articles for classification, while ignoring relevant information from other sources. And based on simple rules or vocabulary, they cannot fully capture the important information in the text.

[0004] At the same time, using a simple similarity calculation method, it is impossible to accurately evaluate the semantic and contextual relationship between keywords. And the lack of processing and transformation methods for similarity calculation results leads to a single classification result.

[0005] Therefore, how to design a news event classification method to provide diversified technical solutions for data acquisition and specific processing methods, and then provide more accurate, comprehensive and explainable news event classification results is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0006] In view of this, the present invention provides a news event classification method, which can capture important information in the text, adopt a multi-level keyword vocabulary and similarity calculation method, can better capture the semantic and contextual relationship between keywords, and can more flexibly determine the classification results.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides a news event classification method, comprising:

[0009] S1. Acquire text data of a news event; the text data includes direct data and indirect data;

[0010] S2, extracting a number of keywords from the text data using a keyword extraction fusion model to obtain keyword phrases;

[0011] S3, based on a multi-level keyword thesaurus, performing multi-stage similarity calculation on the keyword phrase to obtain a similarity calculation result; the multi-level keyword thesaurus includes: a primary thesaurus, a secondary thesaurus and a tertiary thesaurus;

[0012] S4, normalizing the similarity calculation results, and obtaining the probability distribution results of the news events based on a multi-layer perception network model;

[0013] S5. Based on the probability distribution result, the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is compared with a preset probability threshold to obtain a classification result of the news event.

[0014] Among them, in S1, direct data is obtained by using the API interface provided by news websites and data providers; indirect data is obtained by writing a crawler model to access news websites or social media platforms.

[0015] Further, extracting keywords from the text data using a keyword extraction fusion model includes:

[0016] S21, preprocessing the text data, including: word segmentation and removal of stop words;

[0017] S22, based on the preprocessed text data, calculating the term frequency TF and the inverse document frequency IDF of each term to obtain the TF-IDF value of each term;

[0018] S23, taking each term as a node of the graph, using the co-occurrence matrix to calculate the relationship between the terms, and constructing a text graph;

[0019] S24, sorting the text graph using the TextRank algorithm to determine the TextRank score of each term;

[0020] S25, performing weighted fusion of the TF-IDF value and the TextRank score to obtain a comprehensive score for each term;

[0021] S26, sorting based on the comprehensive score of each term, and extracting several keywords through a preset score threshold to obtain a keyword phrase.

[0022] Furthermore, in S3, the multi-level keyword thesaurus is updated based on an intelligent recommendation mechanism.

[0023] Furthermore, in S3, multi-stage similarity calculation is performed on the keyword phrase, including:

[0024] S31. Calculate the similarity between the keyword phrase and each word in the third-level vocabulary:

[0025]

[0026] S32. Calculate the similarity between the keyword phrase and each word in the secondary vocabulary:

[0027]

[0028] S33, calculating the similarity between the keyword phrase and each word in the primary vocabulary:

[0029]

[0030] Among them, c(m,n) represents the keyword phrase vector A m and the word vector B in the secondary vocabulary n distance.

[0031] Furthermore, in S4, the news event classification model includes: an input layer, a hidden layer and an output layer; a ReLU activation function is provided after the hidden layer; and a Softmax activation function is provided after the output layer.

[0032] Furthermore, in S5, based on the probability distribution result, the difference between the probability values ​​of the two news events with the highest probability value and the second highest probability value is compared with a preset probability threshold to obtain a classification result of the news event, including:

[0033] If the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is greater than the preset probability threshold, the news event with the highest probability value is output;

[0034] If the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is less than the preset probability threshold, the two news events with the highest and second highest probability values ​​are output.

[0035] In a second aspect, the present invention provides a news event classification system, comprising:

[0036] Data acquisition module: used to acquire text data of news events; the text data includes: direct data and indirect data;

[0037] Keyword extraction module: used to extract several keywords from the text data using a keyword extraction fusion model to obtain keyword phrases;

[0038] Similarity calculation module: used for performing multi-stage similarity calculation on the keyword phrase based on a multi-level keyword word library to obtain a similarity calculation result; the multi-level keyword word library includes: a primary word library, a secondary word library and a tertiary word library;

[0039] Probability distribution calculation module: used to normalize the similarity calculation results and obtain the probability distribution results of news events based on the multi-layer perception network model;

[0040] Classification result output module: used to compare the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​with a preset probability threshold based on the probability distribution result to obtain the classification result of the news event.

[0041] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned news event classification method when executing the computer program.

[0042] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and the computer program implements the above-mentioned news event classification method when executed by a processor.

[0043] It can be seen from the above technical solutions that, compared with the prior art, the news event classification method provided by the present invention has the following beneficial effects:

[0044] 1. The keyword extraction fusion model can automatically extract several keywords from text data. It can capture important information in the text and convert it into keyword phrases to provide more effective feature representation. Compared with traditional keyword extraction methods based on rules or vocabulary, the fusion model can better adapt to the keyword extraction needs in different fields and contexts.

[0045] 2. The use of multi-level keyword thesaurus and similarity calculation methods can better capture the semantic and contextual relationships between keywords. Through multi-stage similarity calculation, the similarity between keywords can be more accurately evaluated, improving the accuracy and robustness of classification.

[0046] 3. The similarity calculation results are normalized and processed using a multi-layer perception network model, which can map the original data to a probability distribution result. It can effectively handle the differences between different data scales and ranges and convert them into probability distributions, providing more interpretable and comparable classification results.

[0047] 4. By comparing the probability difference between the two news events with the highest and second highest probability values ​​with the preset probability threshold, the classification result can be flexibly determined according to the size of the difference. The accuracy and comprehensiveness of the classification results are taken into account. When the probability difference is large, a definite classification result is given, and when the probability difference is small, more comprehensive information is provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0049] Figure 1 A flow chart of a news event classification method provided by an embodiment of the present invention;

[0050] Figure 2 A flow chart of a method for extracting keywords from text data using a keyword extraction fusion model provided in an embodiment of the present invention;

[0051] Figure 3 A flow chart of a method for calculating multi-stage similarity of keyword phrases provided in an embodiment of the present invention;

[0052] Figure 4 A schematic diagram of a news event classification system provided by an embodiment of the present invention;

[0053] Figure 5 A schematic diagram of the structure of an electronic device is provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] The detection method provided in the embodiment of the present application can be applied to a detection server. The above-mentioned detection server can be hardware or software. When the detection server is hardware, it can be implemented as a distributed server cluster that provides detection services, or it can be implemented as a single server. When the detection server is software, it can be installed in the servers listed above. It can be implemented as multiple software or software modules, or it can be implemented as a single software or software module, which is not specifically limited here.

[0056] Embodiment 1:

[0057] like Figure 1 As shown, this embodiment provides a news event classification method, including:

[0058] S1. Acquire text data of a news event; the text data includes direct data and indirect data;

[0059] S2, extracting a number of keywords from the text data using a keyword extraction fusion model to obtain keyword phrases;

[0060] S3, based on a multi-level keyword thesaurus, performing multi-stage similarity calculation on the keyword phrase to obtain a similarity calculation result; the multi-level keyword thesaurus includes: a primary thesaurus, a secondary thesaurus and a tertiary thesaurus;

[0061] S4, normalizing the similarity calculation results, and obtaining the probability distribution results of the news events based on a multi-layer perception network model;

[0062] S5. Based on the probability distribution result, the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is compared with a preset probability threshold to obtain a classification result of the news event.

[0063] This method can provide more accurate, comprehensive and explainable news event classification results by comprehensively utilizing different types of data, fusion models, multi-level keyword lexicon and similarity calculation, normalization processing and flexible classification result processing.

[0064] The following is a further detailed description of each of the above steps:

[0065] In S1, direct data is obtained by using the API interface provided by news websites and data providers; indirect data is obtained by writing crawler models to access news websites or social media platforms.

[0066] The data acquisition method in this embodiment is not limited to a single specific text. You can use the API interface to obtain direct data and write a crawler model to obtain indirect data. You can use this method to obtain news event data from multiple data sources. Direct data sources provide high-quality, organized data through the API interface, which is suitable for scenarios with high data quality requirements; while indirect data sources can obtain more diverse and comprehensive data through the crawler model, which is suitable for obtaining detailed information on specific websites or social media platforms.

[0067] like Figure 2 As shown, in S2, the keyword extraction fusion model is used to extract keywords from the text data, including:

[0068] S21. Preprocess the text data, including: word segmentation and stop word removal; word segmentation is the process of splitting the text into words, and stop word removal refers to removing common words that appear frequently in the text but are irrelevant to distinguishing the text features. This step helps to convert the text into a discrete sequence of words and reduce redundant information and noise.

[0069] S22. Based on the preprocessed text data, the term frequency TF and inverse document frequency IDF of each term are calculated to obtain the TF-IDF value of each term; TF represents the frequency of a term appearing in the text, and IDF represents the importance of a term in the entire text collection.

[0070] S23. Take each word as a node of the graph, use the co-occurrence matrix to calculate the relationship between words, and construct a text graph; the co-occurrence matrix records the co-occurrence relationship between words in the text. By counting the number of co-occurrences between words, the semantic correlation between them can be revealed.

[0071] S24. Sort the text graph using the TextRank algorithm to determine the TextRank score of each term; assign higher scores to important terms by iteratively calculating the weight transfer between terms. This takes into account the connection relationship and importance of terms in the text graph and can accurately reflect the criticality of terms.

[0072] S25. Weighted fusion of TF-IDF value and TextRank score to obtain a comprehensive score for each term; by weighting TF-IDF value and TextRank score, the frequency and importance of the term in the text can be comprehensively considered to obtain a more comprehensive and accurate term score. This fusion method can balance the advantages of the two methods in keyword extraction and improve the accuracy and diversity of keywords.

[0073] S26, sorting based on the comprehensive score of each term, and extracting a number of keywords through a preset score threshold to obtain a keyword phrase. According to the comprehensive score, the terms are sorted by importance, and then according to the preset score threshold, the terms with scores higher than the threshold are screened out as keywords.

[0074] In this embodiment, the keyword extraction fusion model combines the technical methods of different steps, fully considering factors such as the frequency, importance, and relevance of the terms, so as to extract the most representative and key terms in the text.

[0075] In S3, the multi-level keyword thesaurus is updated based on the intelligent recommendation mechanism.

[0076] In the intelligent recommendation mechanism, the TF-IDF algorithm is used to extract keywords from new data, the semantic similarity between new words and existing vocabulary is calculated through word embedding technology, and a multi-layer classifier is constructed to combine word frequency, context information, similarity and other information to determine the vocabulary level to which the new word should belong.

[0077] At the same time, there is also a user feedback channel, setting up a user feedback interface to allow users to make comments on the accuracy and completeness of the vocabulary. Build an online community to promote communication between users and experts, share the use of the vocabulary, and collect update suggestions.

[0078] The intelligent recommendation and update mechanism of the multi-level keyword database can efficiently and accurately capture emerging vocabulary, timely reflect the latest developments in science and technology, society and other fields, and provide strong support for the intelligent analysis of news events.

[0079] like Figure 3 As shown, in S3, a multi-stage similarity calculation is performed on the keyword phrase, including:

[0080] S31. Calculate the similarity between the keyword phrase and each word in the third-level vocabulary:

[0081]

[0082] S32. Calculate the similarity between the keyword phrase and each word in the secondary vocabulary:

[0083]

[0084] S33, calculating the similarity between the keyword phrase and each word in the primary vocabulary:

[0085]

[0086] Among them, c(m,n) represents the keyword phrase vector A m and the word vector B in the secondary vocabulary n distance.

[0087] For example, the keyword phrases obtained from a news article are: (vector machine, speech-to-text, private cloud, HDFS, remote monitoring)

[0088] The similarity between the keyword phrase and each word in the third-level vocabulary is calculated as follows:

[0089] Supervised learning 0.95, unsupervised learning 0.75; speech recognition 0.95, sentiment analysis 0.55; public cloud services 0.65, hybrid cloud architecture 0.95; data mining 0.65, Hadoop ecosystem 0.85; millimeter wave communication 0.25, network slicing 0.35; smart home 0.35, industrial Internet of Things 0.75;

[0090] The similarity between the keyword phrase and each word in the secondary vocabulary is calculated as follows:

[0091] The similarity between the keyword phrase and each word in the secondary vocabulary is calculated as follows:

[0092] Machine learning based on supervised learning and unsupervised learning is 1.7, natural language processing based on speech recognition and sentiment analysis is 1.5; cloud computing based on public cloud services and hybrid cloud architecture is 1.6, big data based on data mining and Hadoop ecosystem is 1.5; 5G network based on millimeter wave communication and network slicing is 0.6, and Internet of Things based on smart home and industrial Internet of Things is 1.1;

[0093] Calculate the similarity between the keyword phrase and each word in the primary vocabulary:

[0094] Artificial intelligence based on machine learning and natural language processing is 3.2, information technology based on cloud computing and big data is 3.1, and communication technology based on 5G networks and the Internet of Things is 1.6.

[0095] In S4, the original similarity calculation results obtained in S3 are normalized to eliminate the dimension effect and make the comparison between different features more fair and reasonable. The minimum-maximum normalization formula is as follows:

[0096]

[0097] For the given data: artificial intelligence is 3.2, information technology is 3.1, and communication technology is 1.6, the minimum value is 1.6 and the maximum value is 3.2. Therefore, after normalization:

[0098] After AI normalization: 3.2-1.63.2-1.6=1.03.2-1.63.2-1.6=1.0

[0099] Information technology After normalization: 3.1-1.63.2-1.6=0.968753.2-1.63.1-1.6=0.96875

[0100] After normalization of communication technology: 1.6-1.63.2-1.6=0.03.2-1.61.6-1.6=0.0

[0101] Furthermore, the normalized data is used as input features and is passed into a multi-layer perception network model, which includes an input layer, one or more hidden layers, and an output layer.

[0102] The input layer has three nodes, corresponding to the normalized similarity scores of "artificial intelligence", "information technology" and "communication technology". Specifically, there are 3 nodes, a hidden layer is assumed to contain 5 nodes, and the output layer is set according to the predefined number of categories of news events, which is assumed to be C categories, that is, the output layer has C nodes.

[0103] During the training process of the multi-layer perception network model: a large number of pre-labeled news samples are used to train the MLP model, and the samples contain news texts and their category labels.

[0104] The trained MLP model receives the normalized feature vector [1.0, 0.96875, 0.0] as input. After the nonlinear transformation of the hidden layer, it finally generates a C-dimensional probability vector at the output layer, which indicates the probability of the news event belonging to each category. The output is [0.5, 0.45, 0.05]

[0105] In S5, based on the probability distribution result, the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is compared with a preset probability threshold to obtain a classification result of the news event, including:

[0106] If the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is greater than the preset probability threshold, the news event with the highest probability value is output;

[0107] If the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is less than the preset probability threshold, the two news events with the highest and second highest probability values ​​are output.

[0108] By comparing the probability value difference with the preset probability threshold, the certainty of the classification result can be judged according to the size of the difference. When the difference is large, the news event with the highest probability value is output to emphasize certainty and importance; when the difference is small, the two news events with the highest and second highest probability are output to provide more choices and information. This method can provide users with reliable and flexible news event classification results based on the credibility and difference of the probability distribution results.

[0109] If the preset probability threshold is set to 0.1, the final classification result of the news is artificial intelligence and information technology.

[0110] Embodiment 2:

[0111] like Figure 4 As shown, this embodiment provides a news event classification system, including:

[0112] Data acquisition module: acquires text data of news events; text data includes direct data and indirect data;

[0113] Keyword extraction module: uses the keyword extraction fusion model to extract several keywords from the text data and obtain keyword phrases;

[0114] Similarity calculation module: Based on a multi-level keyword word library, multi-stage similarity calculation is performed on the keyword phrase to obtain similarity calculation results; the multi-level keyword word library includes: a primary word library, a secondary word library and a tertiary word library;

[0115] Probability distribution calculation module: normalize the similarity calculation results, use the multi-layer perception network model to process, and obtain the probability distribution results of news events;

[0116] Classification result output module: Based on the probability distribution results, the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is compared with the preset probability threshold to obtain the classification result of the news event.

[0117] Embodiment 3:

[0118] like Figure 5 As shown, this embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the news event classification method of the above embodiment is implemented.

[0119] Embodiment 4:

[0120] This embodiment provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the news event classification method of the above embodiment is implemented.

[0121] The technical solutions in the above embodiments make full use of direct data and indirect data, keyword extraction fusion model, multi-level keyword word library, normalization processing and multi-layer perception network model, as well as classification strategies based on probability value difference and preset probability threshold. These technical features are integrated with each other, making the classification of news events more accurate, comprehensive and flexible, and improving the effect of news event processing.

[0122] In the above embodiments provided in the present application, it should be understood that the disclosed methods, systems, devices and media can be implemented in other ways. The above described methods, systems, devices and media embodiments are only schematic. For example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation. Each functional unit can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0123] The units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0124] The computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. Computer readable media may include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium, etc.

[0125] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A news event classification method, characterized in that: include: S1. Acquire text data of a news event; the text data includes direct data and indirect data; S2, extracting a number of keywords from the text data using a keyword extraction fusion model to obtain keyword phrases; S3, based on a multi-level keyword thesaurus, performing multi-stage similarity calculation on the keyword phrase to obtain a similarity calculation result; the multi-level keyword thesaurus includes: a primary thesaurus, a secondary thesaurus and a tertiary thesaurus; S4, normalizing the similarity calculation results, and obtaining the probability distribution results of the news events based on a multi-layer perception network model; S5. Based on the probability distribution result, the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is compared with a preset probability threshold to obtain a classification result of the news event.

2. A news event classification method according to claim 1, characterized in that: In S1, direct data is obtained by using the API interface provided by news websites and data providers; indirect data is obtained by writing a crawler model to access news websites or social media platforms.

3. A news event classification method according to claim 1, characterized in that: In S2, the keyword extraction fusion model is used to extract keywords from the text data, including: S21, preprocessing the text data, including: word segmentation and removal of stop words; S22, based on the preprocessed text data, calculating the term frequency TF and the inverse document frequency IDF of each term to obtain the TF-IDF value of each term; S23, taking each term as a node of the graph, using the co-occurrence matrix to calculate the relationship between the terms, and constructing a text graph; S24, sorting the text graph using the TextRank algorithm to determine the TextRank score of each term; S25, performing weighted fusion of the TF-IDF value and the TextRank score to obtain a comprehensive score for each term; S26, sorting based on the comprehensive score of each term, and extracting several keywords through a preset score threshold to obtain a keyword phrase.

4. A news event classification method according to claim 3, characterized in that: In the S3, the multi-level keyword thesaurus is updated based on an intelligent recommendation mechanism.

5. A news event classification method according to claim 1, characterized in that: In S3, multi-stage similarity calculation is performed on the keyword phrase, including: S31. Calculate the similarity between the keyword phrase and each word in the third-level vocabulary: S32. Calculate the similarity between the keyword phrase and each word in the secondary vocabulary: S33, calculating the similarity between the keyword phrase and each word in the primary vocabulary: Among them, c(m,n) represents the keyword phrase vector A m and the word vector B in the secondary vocabulary n distance.

6. A news event classification method according to claim 1, characterized in that: In S4, the news event classification model includes: an input layer, a hidden layer and an output layer; a ReLU activation function is set after the hidden layer; and a Softmax activation function is set after the output layer.

7. A news event classification method according to claim 1, characterized in that: In S5, based on the probability distribution result, the difference between the probability values ​​of the two news events with the highest probability value and the second highest probability value is compared with a preset probability threshold to obtain a classification result of the news event, including: If the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is greater than the preset probability threshold, the news event with the highest probability value is output; If the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​is less than the preset probability threshold, the two news events with the highest and second highest probability values ​​are output.

8. A news event classification system, characterized in that: include: Data acquisition module: used to acquire text data of news events; the text data includes: direct data and indirect data; Keyword extraction module: used to extract several keywords from the text data using a keyword extraction fusion model to obtain keyword phrases; Similarity calculation module: used for performing multi-stage similarity calculation on the keyword phrase based on a multi-level keyword word library to obtain a similarity calculation result; the multi-level keyword word library includes: a primary word library, a secondary word library and a tertiary word library; Probability distribution calculation module: used to normalize the similarity calculation results and obtain the probability distribution results of news events based on the multi-layer perception network model; Classification result output module: used to compare the difference between the probability values ​​of the two news events with the highest and second highest probability values ​​with a preset probability threshold based on the probability distribution result to obtain the classification result of the news event.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the news event classification method as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the news event classification method as described in any one of claims 1 to 7 is implemented.

Images