Public opinion data mining methods, systems, electronic devices and storage media

Abstract

This disclosure relates to a method, system, electronic device, and storage medium for public opinion data mining, belonging to the field of public opinion data mining technology. The method includes: acquiring public opinion data and performing attitude analysis on multiple users corresponding to the public opinion data; based on the multiple analysis results corresponding to the attitude analysis, determining the perceptual sensitivity of the users under the same or different cognitions. Embodiments of this disclosure can realize the determination of the perceptual sensitivity of multiple users corresponding to the public opinion data under the same or different cognitions.

Description

Technical Field

[0001] This disclosure relates to the field of public opinion data mining technology, and in particular to a public opinion data mining method, system, electronic device and storage medium. Background Technology

[0002] In the internet age, public opinion data is a collective term for the collection of public opinions and viewpoints. By analyzing the users of public opinion data, we can understand the opinions of the vast majority of netizens. Especially in the business world, it can provide insights into consumer satisfaction with businesses or their products.

[0003] Only those driven by emotion are savages, only those driven by reason are barbarians, and truly free and complete individuals are those who achieve a harmonious balance between reason and emotion. Big data, community building, and value-based marketing have become key factors in the Marketing 4.0 era. For example, the explosive popularity of the Oriental Selection campaign conveyed values ​​related to class, naturalism, and nostalgia, which resonated with the real-life experiences of most consumers, thus boosting brand recognition and sales. Therefore, how to deeply mine public opinion data and determine the perceptual sensitivity of multiple users under the same or different perceptions, in order to empower brand growth and lead new consumer awareness, is a pressing technical challenge. Summary of the Invention

[0004] This disclosure proposes a method, system, electronic device, and storage medium for public opinion data mining.

[0005] According to one aspect of this disclosure, a method for mining public opinion data is provided, including:

[0006] Acquire public opinion data and perform attitude analysis on multiple users corresponding to the public opinion data;

[0007] Based on the multiple analysis results corresponding to the attitude analysis, the perceptual sensitivity of the user under the same cognition or different cognitions is determined.

[0008] Preferably, the method for performing attitude analysis on multiple users corresponding to the public opinion data includes:

[0009] Obtain key information about emotional expression;

[0010] Based on the emotional expression keywords and / or emotional expression emoticons or images in the key information of emotional expression, attitude analysis is performed on multiple users corresponding to the public opinion data.

[0011] Preferably, the method for determining the user's perceptual sensitivity under the same or different cognitions based on multiple analysis results corresponding to the attitude analysis includes:

[0012] Determine the cognition of each of the multiple users;

[0013] Based on the cognition of the multiple users, the multiple users are classified;

[0014] Based on the categorized users, the perceptual sensitivity corresponding to each categorized user is determined.

[0015] Preferably, before determining the perceptual sensitivity of the users under the same or different cognitions, the method for determining the cognitions of the multiple users includes:

[0016] Determine the context and / or plot corresponding to the public opinion data;

[0017] Based on the context and / or plot corresponding to the public opinion data, determine the multiple mainstream values ​​corresponding to the multiple users;

[0018] And configure the top-ranked mainstream values ​​among the multiple mainstream values ​​as the corresponding cognitions of the multiple users; and / or,

[0019] The method for obtaining public opinion data includes:

[0020] Obtain at least one public opinion keyword corresponding to the public opinion data to be mined;

[0021] The public opinion data is obtained by mining the public opinion data based on the aforementioned public opinion keywords; and / or,

[0022] Before performing attitude analysis on the multiple users corresponding to the public opinion data, the method for determining the multiple users corresponding to the public opinion data includes:

[0023] Obtain the social network topology corresponding to the public opinion data, and generate a public opinion propagation search tree based on the social network topology;

[0024] Using the aforementioned public opinion propagation search tree, the initial user of the public opinion data propagation corresponding to the source node of the public opinion propagation search tree is determined;

[0025] Based on the social network topology corresponding to the initial users of the public opinion data, multiple users corresponding to the public opinion data are identified; and / or,

[0026] The perceptual sensitivity is configured as one or more of emotional sensitivity, mood sensitivity, and erotic sensitivity; and / or,

[0027] When the user type is configured as "mother" or the context and / or plot corresponding to the public opinion data is configured as "sole feeding," the cognition or multiple mainstream values ​​are configured as one or more of the following: sense of security, women's care, age pressure, workplace pressure, generational views, views on marriage, traditional aesthetics, anti-traditional aesthetics, and detachment; and / or,

[0028] Before acquiring public opinion data and performing attitude analysis on multiple users corresponding to the public opinion data, a method for determining the public opinion analysis vector corresponding to the public opinion data includes:

[0029] The public opinion data is converted into text data, and the text data is then converted into a word list.

[0030] Using a pre-defined word vector model, each word in the word list is converted into a word vector;

[0031] The word vectors are converted into text vectors for public opinion analysis.

[0032] Based on the aforementioned public opinion analysis text vectors, attitude analysis is performed on multiple users corresponding to the public opinion data.

[0033] Preferably, the method for generating a public opinion propagation search tree based on the social network topology includes:

[0034] A public opinion propagation topology map is generated based on the aforementioned social network topology;

[0035] Based on the aforementioned public opinion propagation topology, a public opinion propagation search tree is constructed; and / or,

[0036] The method for determining the initial user of public opinion corresponding to the source node of the public opinion propagation search tree using the public opinion propagation search tree includes:

[0037] The propagation center value and prior estimate corresponding to the nodes of the public opinion propagation search tree are calculated using the aforementioned public opinion propagation search tree.

[0038] Based on the propagation center value and the prior estimate, the public opinion data is traced to determine the initial user of the public opinion corresponding to the source node of the public opinion propagation search tree; and / or,

[0039] The emotional sensitivity is configured as one or more of the following: hatred, indignation, anxiety, disappointment, sadness, calmness, pleasure, happiness, excitement, and euphoria; and / or,

[0040] The emotional sensitivity is configured as one or more of the following: worry, pain, doubt, surprise, happiness, depression, anxiety, trepidation, and passivity; and / or,

[0041] The erotic sensitivity is configured as one or more of the following: healthy, safe, nutritious, strong, thin, weak, and diseased; and / or,

[0042] Before converting the public opinion data into text data, the method further includes: if the public opinion data is video data, then converting the audio signal corresponding to the video data into text data; and / or,

[0043] The method for converting the audio signal corresponding to the video data into text data includes:

[0044] Determine the number of speakers in the speech signal;

[0045] If the number of speakers is greater than 1, then the identities of the speakers are determined; the start time and end time of each speaker's speech are determined; and the audio signals corresponding to the start time and end time of each speaker's speech are converted into text data.

[0046] Otherwise, the speech signal is directly converted into text data; and / or,

[0047] If the speaker's identity is configured as both a questioner and a responder;

[0048] The text data corresponding to the questioner's identity and the responder's identity are processed to obtain text data that conforms to preset rules; and / or,

[0049] The method for processing the text data corresponding to the questioning identity and the responding identity to obtain text data conforming to preset rules includes:

[0050] Extract the first text features and second text features corresponding to the first text data of the questioning identity and the second text data of the responding identity, respectively;

[0051] Calculate the similarity between the first text feature and the second text feature;

[0052] Based on the aforementioned similarity and a preset similarity, the text data corresponding to the questioner's identity and the responder's identity are processed to obtain text data that conforms to preset rules; and / or,

[0053] The method for processing text data corresponding to the questioner's identity and the responder's identity based on the similarity and a preset similarity to obtain text data conforming to preset rules includes:

[0054] If the similarity is greater than or equal to the preset similarity, then the corresponding first text data and second text data are configured as text data conforming to preset rules; and / or,

[0055] The method for performing attitude analysis on multiple users corresponding to the public opinion data based on the public opinion analysis text vector includes:

[0056] Based on a pre-defined multi-classification model, the sentiment analysis text vectors are used to classify the attitudes of multiple users corresponding to the sentiment data, resulting in multiple analysis results corresponding to the attitude analysis; and / or,

[0057] The method for performing attitude analysis on multiple users corresponding to the public opinion data based on the public opinion analysis text vector further includes:

[0058] Multiple public opinion analysis text vectors corresponding to the public opinion data of the same user on multiple public opinion platforms or on the same public opinion platform are obtained respectively;

[0059] The multiple public opinion analysis text vectors are concatenated and fused separately.

[0060] Based on a preset multi-classification model, the spliced ​​and fused public opinion analysis text vector is used to classify the attitude of each user, resulting in multiple analysis results corresponding to the attitude analysis of each user.

[0061] According to one aspect of this disclosure, a public opinion data mining method is provided, applied to target users, including: the method described above; and,

[0062] Based on the perceptual sensitivity of multiple users corresponding to the public opinion data under the same or different cognitions, target users are recommended.

[0063] According to one aspect of this disclosure, a public opinion data mining method is provided, applied to spokespersons, including: the method described above; and,

[0064] Based on the perceptual sensitivity of multiple users corresponding to the public opinion data under the same or different cognitions, target users are identified; and spokespersons corresponding to the public opinion data are recommended based on the target users.

[0065] According to one aspect of this disclosure, a public opinion data mining system is provided, comprising:

[0066] The analysis unit is used to acquire public opinion data and perform attitude analysis on multiple users corresponding to the public opinion data.

[0067] A determining unit is used to determine the perceptual sensitivity of the user under the same cognition or different cognitions based on multiple analysis results corresponding to the attitude analysis; and / or, a first recommendation unit is used to recommend target users based on the perceptual sensitivity of multiple users corresponding to the public opinion data under the same cognition or different cognitions; and / or, a second recommendation unit is used to recommend spokespersons corresponding to the public opinion data based on the target users.

[0068] According to one aspect of this disclosure, an electronic device is provided, comprising:

[0069] processor;

[0070] Memory used to store processor-executable instructions;

[0071] The processor is configured to execute the aforementioned public opinion data mining method.

[0072] According to one aspect of this disclosure, a computer-readable storage medium is provided that stores computer program instructions thereon, which, when executed by a processor, implement the above-described public opinion data mining method.

[0073] In the embodiments disclosed herein, the proposed public opinion data mining method, system, electronic device, and storage medium aim to solve the current problem that it is impossible to determine the perceptual sensitivity of corresponding users based on public opinion data, thus failing to fundamentally empower brand power growth and lead new consumer perceptions.

[0074] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.

[0075] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0076] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the specification, serve to illustrate the technical solutions of this disclosure.

[0077] Figure 1 A flowchart illustrating a public opinion data mining method according to an embodiment of this disclosure is shown;

[0078] Figure 2 A perceptual sensitivity map corresponding to the user under the same cognition or different cognitions according to an embodiment of the present disclosure is shown;

[0079] Figure 3 This illustration shows multiple mainstream value graphs corresponding to different public opinion data users according to embodiments of the present disclosure;

[0080] Figure 4 This diagram illustrates the content allocation of emotions, moods, desires, situations, and plots according to embodiments of this disclosure.

[0081] Figure 5 A perceptual sensitivity map of the user under the same cognition or different cognitions according to another embodiment of the present disclosure is shown;

[0082] Figure 6 This diagram illustrates a block diagram of a public opinion data mining system according to an embodiment of the present disclosure;

[0083] Figure 7 This is a block diagram illustrating an electronic device 800 according to an exemplary embodiment;

[0084] Figure 8 This is a block diagram illustrating an electronic device 1900 according to an exemplary embodiment. Detailed Implementation

[0085] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0086] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0087] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0088] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0089] It is understood that the various method embodiments mentioned above in this disclosure can be combined with each other to form combined embodiments without violating the principle and logic. Due to space limitations, this disclosure will not elaborate further.

[0090] In addition, this disclosure also provides a public opinion data mining system, electronic equipment, computer-readable storage medium, and program, all of which can be used to implement any of the public opinion data mining methods provided in this disclosure. The corresponding technical solutions and descriptions are described in the relevant section of the method and will not be repeated here.

[0091] Figure 1 A flowchart illustrating a public opinion data mining method according to an embodiment of this disclosure is shown, such as... Figure 1 As shown, the public opinion data mining method includes: Step S101: acquiring public opinion data and performing attitude analysis on multiple users corresponding to the public opinion data; Step S102: determining the perceptual sensitivity of the users under the same or different cognitions based on the multiple analysis results corresponding to the attitude analysis. This determines the perceptual sensitivity of multiple users corresponding to the public opinion data under the same or different cognitions, thereby empowering brand power growth and leading new consumer perceptions.

[0092] Step S101: Obtain public opinion data and perform attitude analysis on multiple users corresponding to the public opinion data.

[0093] In embodiments of this disclosure, the method for performing attitude analysis on multiple users corresponding to the public opinion data includes: obtaining key information on emotional expression; and performing attitude analysis on multiple users corresponding to the public opinion data based on emotional expression keywords and / or emotional expression emoticons or images in the key information on emotional expression.

[0094] In the embodiments of this disclosure and other possible embodiments, those skilled in the art can configure the emotional expression keywords in the key information of emotional expression according to actual needs. For example, the emotional expression keywords can be configured as words or phrases corresponding to wry smile, unhappy, grinning, infatuated, sticking out one eye and tongue, shy, and / or smiling; or, the emotional expression emoticons or images can be configured as emoticons or images corresponding to wry smile, unhappy, grinning, infatuated, sticking out one eye and tongue, shy, and / or smiling.

[0095] In the embodiments of this disclosure, the method for obtaining public opinion data includes: obtaining at least one public opinion keyword corresponding to the public opinion data to be mined; and mining the public opinion data to be mined based on the public opinion keyword to obtain the public opinion data.

[0096] In the embodiments disclosed herein and other possible embodiments, those skilled in the art can configure the public opinion keywords according to actual needs. For example, the names of people or times corresponding to the public opinion, or other possible public opinion keywords.

[0097] In the embodiments of this disclosure and other possible embodiments, the public opinion data to be mined is mined from at least one designated public opinion platform based on the public opinion keywords to obtain the public opinion data. Those skilled in the art can configure the public opinion platform according to actual needs. For example, the public opinion platform can be configured as one or more of the following: official Weibo accounts, WeChat public accounts, portal websites and government websites, WeChat, Weibo, forums, Tieba and news media websites, Toutiao, Baijiahao, WeChat public accounts and other self-media accounts, Douyin, Kuaishou, Zhihu, etc.

[0098] In embodiments of this disclosure, before determining the perceptual sensitivity of the users under the same or different cognitions, a method for determining the cognitions corresponding to the multiple users includes: determining the context and / or plot corresponding to the public opinion data; determining multiple mainstream values ​​corresponding to the multiple users based on the context and / or plot corresponding to the public opinion data; and configuring the multiple mainstream values ​​with the highest ranking among the multiple mainstream values ​​as the cognitions corresponding to the multiple users. For example, the multiple mainstream values ​​can be configured as the three mainstream values ​​with the highest ranking in sequence.

[0099] In the embodiments disclosed herein and other possible embodiments, those skilled in the art can configure the context and / or plot corresponding to the public opinion data according to actual needs. For example, the plot can be configured as one or more of the following: feeling needed, doubting the true meaning of marriage, life being difficult (being cursed in life, family in chaos), and searching for oneself; the mood can be one or more of the following: learning knowledge, feeding alone, waking up in fright, being unable to move, exercising, insomnia, sighing (sighing while touching one's stomach), psychotherapy, working or studying continuously (working non-stop), lack of experience, staying up late, drinking water, conflicts between mother-in-law and daughter-in-law in raising children, regulating breathing, massage, and regulating the body.

[0100] Figure 2 This illustrates a perceptual sensitivity map of the user under the same or different cognitive states according to embodiments of this disclosure. Figure 2 As shown, for different public opinion data, the perceptual sensitivity corresponding to the user under the same or different cognitions is given. For example, if the public opinion data is configured as Xie Guangkun selling his grandson in "Country Love", the perceptual sensitivity corresponding to wealth-oriented thinking is aversion. Here, n / a represents empty (no public opinion data). Under the same underlying cognition, consumers (multiple users corresponding to the public opinion data) will express significant differences in their responses to different events (the public opinion data). By deconstructing the underlying cognition, the perceptual sensitivity of users can be located... Figure 2 This allows brands to intuitively see the high / low perception sensitivity of consumers (multiple users corresponding to the aforementioned public opinion data). Simultaneously, it can serve as a content / script / scenario expansion library for new media operations.

[0101] Figure 3 This illustrates multiple mainstream value graphs corresponding to different public opinion data users according to embodiments of this disclosure. For example... Figure 3 As shown, the charts present multiple mainstream value graphs corresponding to user sentiment data in the smart electronics industry, the beauty industry, and the video beverage industry. The data for each of the top-ranked mainstream values ​​is 12.

[0102] Figure 4This diagram illustrates the content allocation of emotions, feelings, desires, situations, and plots according to embodiments of this disclosure. For example... Figure 4 As shown in the embodiments of this disclosure, the perceptual sensitivity is configured as one or more of emotional sensitivity, mood sensitivity, and erotic sensitivity. Specifically, the emotional sensitivity is configured as one or more of hatred, indignation, anxiety, disappointment, sadness, calmness, pleasure, happiness, excitement, and exhilaration; and / or, the mood sensitivity is configured as one or more of worry, pain, doubt, surprise, joy, depression, unease, trepidation, and passivity; and / or, the erotic sensitivity is configured as one or more of health, safety, nutrition, strength, thinness, weakness, and illness. The scenario can be configured as one or more of feeling needed, doubting the true meaning of marriage, life being difficult (being cursed in life, domestic chaos), and searching for oneself; the emotional mirror can be one or more of learning knowledge, feeding alone, waking up in fright, being unable to move, exercising, insomnia, sighing (sighing while touching the stomach), psychotherapy, working or studying continuously (working non-stop), lack of experience, staying up late, drinking water, conflicts between mother-in-law and daughter-in-law regarding childcare, regulating breathing, massage, and regulating the body.

[0103] In the embodiments of this disclosure and other possible embodiments, the emotional sensitivity is configured as one or more of hatred, indignation, anxiety, disappointment, sadness, calmness, pleasure, happiness, excitement, and agitation; wherein the emotional sensitivity corresponding to hatred, indignation, anxiety, disappointment, sadness, calmness, pleasure, happiness, excitement, and agitation increases or decreases in sequence.

[0104] In the embodiments of this disclosure and other possible embodiments, those skilled in the art can configure the emotions, feelings, desires, situations, and plots according to actual needs. For example, the emotional sensitivity can also be configured as one or more of the following: respect, transcendence, empathy, pride, helplessness, loneliness, loss, frustration, and confusion; the emotional sensitivity can also be configured as one or more of the following: joy, anger, sadness, and surprise; the desire sensitivity can also be configured as one or more of the following: survival, job security, friendship, and confidence; the plot can also be configured as one or more of the following: a child from an ordinary family achieving success through knowledge-based e-commerce, teachers becoming popular through emotional live streams, someone's mid-life comeback, a live stream reminiscing about rural life and parents and relatives, and a stock price rebounding after a limit-down; the situation can also be configured as one or more of the following: being forced into a live stream, bursting into tears while listening to someone's story, being touched by a similar experience, or placing an order for past sentiments.

[0105] In embodiments of this disclosure, before converting the public opinion data into text data, the method further includes: if the public opinion data is video data, then converting the audio signal corresponding to the video data into text data.

[0106] In an embodiment of this disclosure, the method for converting the audio signal corresponding to the video data into text data includes: determining the number of speakers in the audio signal; if the number of speakers is greater than 1, determining the identity corresponding to each speaker; determining the speaking start time and speaking end time of each identity; and converting the audio signal corresponding to the speaking start time and speaking end time of each identity into text data; otherwise, directly converting the audio signal into text data.

[0107] In embodiments of this disclosure and other possible embodiments, the method for determining the number of speakers in the speech signal includes: identifying the speakers of the speech signal using speaker recognition technology; and determining the number of speakers in the speech signal based on the speakers.

[0108] In embodiments of this disclosure and other possible embodiments, the method for determining the speaking start time and speaking end time of each of the identities includes: using voice endpoint detection technology to determine the speaking start time and speaking end time of each of the identities.

[0109] In embodiments of this disclosure and other possible embodiments, the method for converting the audio signals corresponding to the speech start time and speech end time of each of the identities into text data includes: using speech recognition technology to convert the audio signals corresponding to the speech start time and speech end time of each of the identities into text data.

[0110] In the embodiments of this disclosure, if the speaker's identity is configured as a questioner and a responder, the text data corresponding to the questioner and the responder are processed to obtain text data that conforms to preset rules.

[0111] In embodiments of this disclosure, the method for processing the text data corresponding to the questioning identity and the responding identity to obtain text data conforming to preset rules includes: extracting first text features and second text features corresponding to the first text data of the questioning identity and the second text data of the responding identity, respectively; calculating the similarity between the first text features and the second text features; and processing the text data corresponding to the questioning identity and the responding identity based on the similarity and the preset similarity to obtain text data conforming to preset rules.

[0112] In embodiments of this disclosure, the method for processing text data corresponding to the questioner's identity and the responder's identity based on the similarity and a preset similarity to obtain text data conforming to preset rules includes: if the similarity is greater than or equal to the preset similarity, then configuring the corresponding first text data and second text data as text data conforming to the preset rules. Those skilled in the art can configure the value corresponding to the preset similarity according to actual needs, and the similarity can be configured as cosine similarity.

[0113] In embodiments of this disclosure and other possible embodiments, the method for extracting the first text data corresponding to the questioner's identity and the second text data corresponding to the responder's identity includes: obtaining, based on a preset word vector model, the first text data and the second text data corresponding to the first text data and the second text data respectively; calculating the sum of the first word vectors and the sum of the second word vectors corresponding to the first key feature word vectors and the second key feature word vectors respectively; calculating the first length and the second length of the first text data and the second text data respectively; determining the first text feature corresponding to the first text data of the questioner's identity based on the sum of the first word vectors and the first length; and determining the second text feature corresponding to the second text data of the questioner's identity based on the sum of the second word vectors and the second length.

[0114] In embodiments of this disclosure and other possible embodiments, the method for determining the first text feature corresponding to the first text data of the questioner's identity based on the sum of the first word vectors and the first length includes: dividing the sum of the first word vectors by the first length to obtain the first text feature corresponding to the first text data of the questioner's identity.

[0115] In embodiments of this disclosure and other possible embodiments, the method for determining the second text feature corresponding to the second text data of the question identity based on the sum of the second word vectors and the second length includes: dividing the sum of the second word vectors by the second length to obtain the second text feature corresponding to the second text data of the question identity.

[0116] Step S102: Based on the multiple analysis results corresponding to the attitude analysis, determine the user's perceptual sensitivity under the same cognition or different cognitions.

[0117] In embodiments of this disclosure, the method for determining the perceptual sensitivity of a user under the same or different cognitions based on multiple analysis results corresponding to the attitude analysis includes: determining the cognitions corresponding to the multiple users respectively; classifying the multiple users according to the cognitions corresponding to the multiple users; and determining the perceptual sensitivity of the classified users respectively based on the classified users.

[0118] In embodiments of this disclosure, when the types of the multiple users are configured as mothers (e.g., new mothers) or the context and / or plot corresponding to the public opinion data are configured as feeding alone, the cognition or multiple mainstream values ​​are configured as one or more of the following: sense of security, women's care, age pressure, workplace pressure, generational views, views on marriage, traditional aesthetics, anti-traditional aesthetics, and detachment.

[0119] Figure 5 A perceptual sensitivity map of the user under the same or different cognitions is shown according to another embodiment of this disclosure. For example... Figure 5 As shown, the public opinion data configuration indicates that when new mothers are reassigned or dismissed, the perceived sensitivity of users under the corresponding cognition of workplace pressure is indignation. For example, multiple users are configured as new mothers, and the cognitions or mainstream values ​​corresponding to these multiple users (new mothers) are determined. For example, the cognitions or mainstream values ​​corresponding to these multiple users (new mothers) include: sense of security, women's care, age pressure, workplace pressure, generational views, views on marriage, traditional aesthetics, anti-traditional aesthetics, and one or more of detachment. Based on the cognitions or mainstream values ​​corresponding to these multiple users, the multiple users (new mothers) are classified to obtain users (new mothers) classified under different cognitions or mainstream values, and the perceived sensitivity of each classified user (new mother) is determined.

[0120] In an embodiment of this disclosure, before acquiring public opinion data and performing attitude analysis on multiple users corresponding to the public opinion data, a method for determining the public opinion analysis vector corresponding to the public opinion data includes: converting the acquired public opinion data into text data and converting the text data into a word list; using a preset word vector model, converting each word in the word list into a word vector; converting the word vectors into public opinion analysis text vectors; and performing attitude analysis on multiple users corresponding to the public opinion data based on the public opinion analysis text vectors.

[0121] In embodiments of this disclosure and other possible embodiments, the preset word vector model can be configured as one or more of the word2vec model or the GloVe model. The training method for the preset word vector model includes: obtaining training corpus for the preset word vector model based on the text data; inputting the training corpus into the preset vector model to train the preset word vector model; and using the trained preset word vector model to convert each word in the word list into a word vector.

[0122] In embodiments of this disclosure and other possible embodiments, a method for obtaining training corpus for a text data word vector model based on the text data includes: segmenting the text data into words, converting the text data into a word list, and obtaining training corpus for the word vector model.

[0123] In embodiments of this disclosure and other possible embodiments, the method for converting the text data into a word list includes: segmenting the text using a word segmentation tool (e.g., jieba word segmentation tool) and a custom dictionary within the word segmentation tool; setting a stop word list; filtering the text data using the stop word list; and converting the text data into a word list. The stop words can be selected by adding words to be removed from an existing stop word list according to specific scenarios. Stop words can be configured as one or more of modal particles, auxiliary words, and / or punctuation marks.

[0124] In embodiments of this disclosure and other possible embodiments, the method of converting the word vectors into public opinion analysis text vectors includes: concatenating the word vectors to obtain the public opinion analysis text vectors. Specifically, the method of concatenating the word vectors to obtain the public opinion analysis text vectors includes: determining the mean word vector, maximum word vector, and minimum word vector corresponding to each word vector; and processing the mean word vector, maximum word vector, and minimum word vector...

[0125] In the embodiments of this disclosure, the method for performing attitude analysis on multiple users corresponding to the public opinion data based on the public opinion analysis text vector includes classifying the attitudes of multiple users corresponding to the public opinion data using the public opinion analysis text vector based on a preset multi-classification model, and obtaining multiple analysis results corresponding to the attitude analysis.

[0126] In embodiments of this disclosure and other possible embodiments, the preset multi-classification model can be configured as a machine learning (ML) classification model. For example, it can be configured as one or more of support vector machines, decision trees, random forests, K-nearest neighbors, logistic regression, adaptive augmentation, linear discriminant analysis, and multilayer perceptrons.

[0127] In embodiments of this disclosure and other possible embodiments, the preset multi-classification model is configured as a Random Forest (RF). RF constructs multiple decision trees and comprehensively evaluates the predictions of the multiple decision trees to obtain multiple analysis results corresponding to the attitude analysis.

[0128] For example, in the embodiments of this disclosure and other possible embodiments, the preset multi-classification model is configured as a model corresponding to the K-nearest Neighbor (KNN) algorithm, thereby classifying the attitudes of multiple users corresponding to the public opinion data and obtaining multiple analysis results corresponding to the attitude analysis.

[0129] The core idea of ​​KNN is that if a sample's k nearest neighbors in the feature space mostly belong to a certain class, then the sample also belongs to that class and possesses the characteristics of samples in that class. Specifically, because the KNN model is a non-parametric, lazy algorithm, it means that it makes no assumptions about the data features; that is, the model structure built by KNN is determined by the data features. Therefore, the KNN model does not require a large amount of data for training, offering advantages such as fast training speed and insensitivity to outliers, but disadvantages such as high memory requirements and sensitivity to the size of inconsistent core data features.

[0130] In addition, the KNN model is also affected by several factors, namely the method of calculating the distance between points, the selection of the value of K, and the decision rule.

[0131] (a) Point distance calculation

[0132] Common methods for measuring distances between points in space include Manhattan distance calculation and Euclidean distance calculation. However, the KNN model usually uses Euclidean distance, and the mathematical expression for the Euclidean distance between two points in two-dimensional space is shown in equation (1).

[0133]

[0134] Where ρ is the Euclidean distance between two points (x1, y1) and (x2, y2) in two-dimensional space.

[0135] For three-dimensional space, the mathematical expression for Euclidean distance is shown in equation (2).

[0136]

[0137] Where ρ is the Euclidean distance between two points (x1, y1, z1) and (x2, y2, z2) in three-dimensional space.

[0138] For n-dimensional space, the mathematical expression for this distance is shown in equation (3).

[0139]

[0140] Where ρ represents (x1,y1), (x2,y2)...(x) in n-dimensional space. n ,y n The Euclidean distance between them.

[0141] (b) Selection of K value

[0142] The value of K is crucial to the accuracy of the model. Typically, a small K value is chosen first, and then the value of K is gradually increased. The variance of the validation set is then calculated to find a suitable K value. When K is increased, the error rate generally decreases initially because more surrounding samples are available for reference, leading to improved classification performance. However, the error rate increases with even larger K values.

[0143] (c) Selection of decision rules

[0144] In classification tasks, KNN models typically employ either majority voting or weighted majority voting. In majority voting, each sample has the same weight, while in weighted majority voting, each sample has a different weight, usually calculated by inversely proportional to the distance. In regression tasks, KNN models generally use either the average method or a weighted average method.

[0145] In the embodiments of this disclosure, the method for performing attitude analysis on multiple users corresponding to the public opinion data based on the public opinion analysis text vectors further includes: obtaining multiple public opinion analysis text vectors corresponding to the public opinion data of the same user on multiple public opinion platforms or on the same public opinion platform; concatenating and fusing the multiple public opinion analysis text vectors; and, based on a preset multi-classification model, using the concatenated and fused public opinion analysis text vectors to classify the attitude of each user, thereby obtaining multiple analysis results corresponding to the attitude analysis of each user.

[0146] In embodiments of this disclosure, before performing attitude analysis on the multiple users corresponding to the public opinion data, the method for determining the multiple users corresponding to the public opinion data includes: obtaining the social network topology corresponding to the public opinion data, and generating a public opinion propagation search tree based on the social network topology; using the public opinion propagation search tree to determine the initial user of public opinion corresponding to the source node of the public opinion propagation search tree of the public opinion data; and determining the multiple users corresponding to the public opinion data based on the social network topology corresponding to the initial user of public opinion.

[0147] In embodiments of this disclosure and other possible embodiments, the method for determining multiple users corresponding to the public opinion data based on the social network topology corresponding to the initial user of the public opinion includes: determining multiple first propagation nodes connected to the initial node of the social network topology corresponding to the initial user of the public opinion; determining second propagation nodes connected to the multiple first propagation nodes based on the social network topology; until, determining a final propagation node connected to the previous level propagation node based on the social network topology; and determining the initial node and the propagation nodes as multiple users corresponding to the public opinion data.

[0148] In embodiments of this disclosure, the method for generating a public opinion propagation search tree based on the social network topology includes: generating a public opinion propagation topology map based on the social network topology; and constructing a public opinion propagation search tree based on the public opinion propagation topology map.

[0149] In embodiments of this disclosure, the method for determining the initial user of public opinion corresponding to the source node of the public opinion propagation search tree using the public opinion propagation search tree includes: calculating the propagation center value and prior estimate value corresponding to the node of the public opinion propagation search tree using the public opinion propagation search tree; tracing the source of the public opinion data based on the propagation center value and the prior estimate value to determine the initial user of public opinion corresponding to the source node of the public opinion propagation search tree of the public opinion data.

[0150] In embodiments of this disclosure and other possible embodiments, before obtaining the social network topology corresponding to the public opinion data, establishing the social network topology corresponding to the public opinion data includes: obtaining users under preset information and address information; and establishing the social network topology corresponding to the public opinion data based on the preset information. The user's address information can be configured as the user's IP address information, and the preset information can be configured as webpage information corresponding to the user's browsing address.

[0151] In the embodiments of this disclosure and other possible embodiments, the method for establishing a social network topology corresponding to the public opinion data based on the address information and preset information further optimizes the users for constructing the social network topology, including: using the address information of the users to determine potential users; and using the preset information to select the potential users to obtain the final users for constructing the social network topology.

[0152] In the embodiments of this disclosure and other possible embodiments, in the social network topology, users can be abstracted as different public opinion propagation search tree nodes, and the interaction between users represents the connection between two public opinion propagation search tree nodes. When public opinion data spreads in the social network topology, the source node of the public opinion propagation search tree can be quickly located using a source tracing algorithm, minimizing the negative impact of public opinion information on the network. Therefore, when performing source tracing calculations, a propagation network needs to be constructed based on the structure of the social network topology and the set of public opinion propagation search tree nodes being propagated. The source tracing algorithm is then applied to the propagation network to quickly and accurately locate the source node of the public opinion propagation search tree. In the social network topology, each edge connects two public opinion propagation search tree nodes, and the edge relationships of the social network topology can form an adjacency matrix.

[0153] In the embodiments disclosed herein and other possible embodiments, for a piece of public opinion data, a public opinion propagation search tree node will be in one of two states: "believe" or "disbelieve." "Believe" means that the public opinion propagation search tree node has been propagated; "disbelieve" means that the public opinion propagation search tree node has not been propagated and is in a susceptible state. When a piece of public opinion information spreads in the social network topology, the public opinion propagation search tree node in the propagation state will propagate to the public opinion propagation search tree nodes around it, forming a local propagation network after a certain time step. Therefore, the public opinion propagation search tree node initially in the propagation state is the core public opinion propagation search tree node of the propagation network. Other public opinion propagation search tree nodes in the propagation network are all caused by the propagation of this core public opinion propagation search tree node, so the importance of this core public opinion propagation search tree node in the network is higher than that of other propagating public opinion propagation search tree nodes. If a public opinion propagation search tree node is propagated, then that public opinion propagation search tree node will remain in the propagation state and will not be propagated again.

[0154] In the embodiments of this disclosure and other possible embodiments, a set of propagated public opinion propagation search tree nodes is constructed within a certain propagation round. If a public opinion propagation search tree node is propagated, it is added to the set of propagating public opinion propagation search tree nodes. When information propagation stops, all propagated public opinion propagation search tree nodes and the public opinion propagation topology graph are obtained based on the set of propagating public opinion propagation search tree nodes and the adjacency matrix.

[0155] In embodiments of this disclosure and other possible embodiments, the method for generating a public opinion propagation topology graph based on the social network topology includes: configuring propagation probabilities for the public opinion propagation search tree nodes of the social network topology; determining whether the public opinion propagation search tree nodes of the social network topology are propagated based on a set propagation round; if so, determining the propagated public opinion propagation search tree nodes as propagating public opinion propagation search tree nodes, and generating a public opinion propagation topology graph based on the propagating public opinion propagation search tree node set and the corresponding adjacency matrix. The propagation probability can be configured as the propagation probability corresponding to an easy propagation state, a difficult-to-propagate state, or a difficult-to-propagate state. Furthermore, based on the public opinion propagation topology graph induced in the social network topology, a breadth-first public opinion propagation search tree with the root public opinion propagation search tree nodes is constructed based on the public opinion propagation topology graph; then, a source tracing operation is performed to determine the initial user of the public opinion data propagation corresponding to the source node of the public opinion propagation search tree.

[0156] In embodiments of this disclosure and other possible embodiments, the method for constructing a public opinion propagation search tree based on the public opinion propagation topology graph includes: determining the root public opinion propagation search tree node of the public opinion propagation topology graph, and establishing a public opinion propagation search tree based on the root public opinion propagation search tree node. In embodiments of this disclosure and other possible embodiments, establishing a public opinion propagation search tree based on the root public opinion propagation search tree node can be a breadth-first public opinion propagation search tree established based on the root public opinion propagation search tree node.

[0157] In embodiments of this disclosure and other possible embodiments, the method for calculating the propagation center value and the prior estimate value corresponding to the public opinion propagation search tree node using the public opinion propagation search tree includes: calculating a first probability value and a second probability value corresponding to the public opinion propagation topology graph under the public opinion propagation search tree node based on the public opinion propagation search tree; and configuring the first probability value and the second probability value as the propagation center value and the prior estimate value corresponding to the public opinion propagation search tree node, respectively.

[0158] In embodiments of this disclosure and other possible embodiments, the method of tracing the source of public opinion data based on the propagation center value and the prior estimate to determine the source node of the public opinion data propagation search tree includes: determining the tracing probability of the public opinion propagation search tree node based on the propagation center value and the prior estimate; tracing the source of the public opinion data based on the tracing probability and a set tracing probability to determine the source node of the public opinion data propagation search tree. Those skilled in the art can set the set tracing probability according to actual needs. In embodiments of this disclosure and other possible embodiments, the method of tracing the source of public opinion data based on the tracing probability and a set tracing probability to determine the source node of the public opinion data propagation search tree includes: if the tracing probability is greater than or equal to the set tracing probability, then determining the public opinion propagation search tree node corresponding to the set tracing probability as the source node of the public opinion data propagation search tree; otherwise, determining the public opinion propagation search tree node corresponding to the set tracing probability as a non-public opinion propagation search tree source node of the public opinion data propagation.

[0159] In an embodiment of this disclosure, the method for determining the source probability of the public opinion propagation search tree node based on the propagation center value and the prior estimate includes: multiplying the propagation center value by the prior estimate to determine the source probability of the public opinion propagation search tree node.

[0160] The embodiments of this disclosure also propose a public opinion data mining method applied to target users, including: as described above, recommending target users based on the perceptual sensitivity of multiple users corresponding to the public opinion data under the same or different cognitions.

[0161] The embodiments of this disclosure also propose a public opinion data mining method applied to spokespersons, including: as described above, determining target users based on the perceptual sensitivity of multiple users corresponding to the public opinion data under the same or different cognitions; and recommending spokespersons corresponding to the public opinion data based on the target users.

[0162] The execution entity of a public opinion data mining method can be a public opinion data mining system. For example, the method can be executed by a terminal device, server, or other processing device. The terminal device can be a user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, in-vehicle device, wearable device, etc. In some possible implementations, the public opinion data mining method can be implemented by a processor calling computer-readable instructions stored in memory.

[0163] Those skilled in the art will understand that, in the above-described public opinion data mining method of specific implementation, the writing order of each step does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0164] Figure 6 A block diagram of a public opinion data mining system according to an embodiment of this disclosure is shown. Figure 6 As shown, the public opinion data mining system includes: an analysis unit, used to acquire public opinion data and perform attitude analysis on multiple users corresponding to the public opinion data; and a determination unit, used to determine the perceptual sensitivity of the user under the same cognition or different cognitions based on the multiple analysis results corresponding to the attitude analysis.

[0165] like Figure 6 As shown, the public opinion data mining system, applied to target users, includes: an analysis unit for acquiring public opinion data and performing attitude analysis on multiple users corresponding to the public opinion data; a determination unit for determining the perceptual sensitivity of the user under the same or different cognitions based on the multiple analysis results corresponding to the attitude analysis; and a first recommendation unit for recommending target users based on the perceptual sensitivity of multiple users corresponding to the public opinion data under the same or different cognitions.

[0166] like Figure 6 As shown, the public opinion data mining system, applied to target users, includes: an analysis unit for acquiring public opinion data and performing attitude analysis on multiple users corresponding to the public opinion data; a determination unit for determining the perceptual sensitivity of the user under the same or different cognitions based on the multiple analysis results corresponding to the attitude analysis; a second recommendation unit for recommending target users based on the perceptual sensitivity of multiple users corresponding to the public opinion data under the same or different cognitions; and recommending spokespersons corresponding to the public opinion data based on the target users.

[0167] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to execute the public opinion data mining method described in the above method embodiments. The specific implementation can be referred to the description of the above public opinion data mining method embodiments, which will not be repeated here for the sake of brevity.

[0168] This disclosure also proposes a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the aforementioned public opinion data mining method. The computer-readable storage medium may be a non-volatile computer-readable storage medium.

[0169] This disclosure also proposes an electronic device, including: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured for the aforementioned public opinion data mining method. The electronic device may be provided as a terminal, a server, or other form of device.

[0170] Figure 7 This is a block diagram illustrating an electronic device 800 according to an exemplary embodiment. For example, the electronic device 800 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, or other terminal.

[0171] Reference Figure 7 The electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power supply component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.

[0172] Processing component 802 typically controls the overall operation of electronic device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the aforementioned public opinion data mining method. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

[0173] Memory 804 is configured to store various types of data to support the operation of electronic device 800. Examples of this data include instructions for any application or method operating on electronic device 800, contact data, phonebook data, messages, pictures, videos, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0174] Power supply component 806 provides power to various components of electronic device 800. Power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 800.

[0175] Multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the electronic device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0176] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when electronic device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0177] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0178] Sensor assembly 814 includes one or more sensors for providing state assessments of various aspects of electronic device 800. For example, sensor assembly 814 can detect the on / off state of electronic device 800, the relative positioning of components such as the display and keypad of electronic device 800, changes in position of electronic device 800 or a component of electronic device 800, the presence or absence of user contact with electronic device 800, orientation or acceleration / deceleration of electronic device 800, and temperature changes of electronic device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0179] Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices. Electronic device 800 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0180] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0181] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 804 including computer program instructions that can be executed by a processor 820 of an electronic device 800 to perform the above-described method.

[0182] Figure 8 This is a block diagram illustrating an electronic device 1900 according to an exemplary embodiment. For example, the electronic device 1900 may be provided as a server. (Refer to...) Figure 8 The electronic device 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by memory 1932 for storing instructions, such as application programs, that can be executed by the processing component 1922. The application programs stored in memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1922 is configured to execute instructions to perform the methods described above.

[0183] Electronic device 1900 may also include a power supply component 1926 configured to perform power management of electronic device 1900, a wired or wireless network interface 1950 configured to connect electronic device 1900 to a network, and an input / output (I / O) interface 1958. Electronic device 1900 can operate on an operating system stored in memory 1932, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or similar.

[0184] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions that can be executed by a processing component 1922 of an electronic device 1900 to perform the above-described method.

[0185] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0186] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0187] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0188] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0189] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0190] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0191] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0192] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0193] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A public opinion data mining method, characterized in that, The method comprises the following steps: Obtaining public opinion data, comprising: obtaining at least one public opinion keyword corresponding to the public opinion data to be mined; and mining the public opinion data to be mined according to the public opinion keyword to obtain public opinion data; Generating a public opinion propagation search tree based on the social network topology corresponding to the public opinion data; calculating the propagation center value and the prior estimate value corresponding to the public opinion propagation search tree node by using the public opinion propagation search tree; tracing the public opinion data according to the propagation center value and the prior estimate value to determine the public opinion initial user corresponding to the public opinion propagation search tree source node of the public opinion data propagation; determining a plurality of users corresponding to the public opinion data based on the social network topology corresponding to the public opinion initial user; converting the public opinion data into text data, and converting the text data into a word list; converting each word in the word list into a word vector by using a preset word vector model; converting the word vector into a public opinion analysis text vector; performing attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector; wherein the attitude analysis on the plurality of users corresponding to the public opinion data comprises: obtaining emotional expression key information; performing attitude analysis on the plurality of users corresponding to the public opinion data based on the emotional expression key information; the conversion of the word vector into the public opinion analysis text vector comprises: splicing the word vector to obtain the public opinion analysis text vector; The attitude analysis based on the public opinion analysis text vector comprises: performing attitude classification on the plurality of users corresponding to the public opinion data by using the public opinion analysis text vector based on a preset multi-classification model to obtain a plurality of analysis results corresponding to the attitude analysis; the attitude analysis based on the public opinion analysis text vector further comprises: obtaining a plurality of public opinion analysis text vectors corresponding to the public opinion data of the same user on a plurality of public opinion platforms or the same public opinion platform; splicing and fusing the plurality of public opinion analysis text vectors; performing attitude classification on each user by using the spliced and fused fusion public opinion analysis text vector based on a preset multi-classification model to obtain a plurality of analysis results corresponding to the attitude analysis of each user; Determining the perception sensitivity of the user under the same cognition or different cognition based on the plurality of analysis results corresponding to the attitude analysis comprises: determining the situation or scenario corresponding to the public opinion data; determining a plurality of mainstream values corresponding to the plurality of users based on the situation or scenario corresponding to the public opinion data; configuring a plurality of mainstream values ranked highest in turn in the plurality of mainstream values as the cognition corresponding to the plurality of users; classifying the plurality of users according to the cognition corresponding to the plurality of users; and determining the perception sensitivity corresponding to the classified user based on the classified user. 2.The public opinion data mining method of claim 1, wherein, The perception sensitivity is configured as emotional sensitivity, emotional sensitivity, and emotional sensitivity. The emotional sensitivity is configured as one or more of loathing, indignation, anxiety, disappointment, sadness, calmness, joy, happiness, excitement, and excitement. The emotional sensitivity is configured as one or more of worry, pain, doubt, surprise, happiness, depression, unease, and passivity. The emotional sensitivity is configured as one or more of health, safety, nutrition, strong, weak, weak, and disease. 3.The public opinion data mining method according to any one of claims 1 or 2, characterized in that, When the type of the plurality of users is configured as a mother or the situation or scenario corresponding to the public opinion data is configured as feeding alone, the plurality of mainstream values is configured as safety, female care, age pressure, job pressure, intergenerational view, marriage view, traditional aesthetic, anti-traditional aesthetic, and transcendentalism.

4. The public opinion data mining method according to any one of claims 1 or 2, characterized in that, The public opinion propagation search tree is generated based on the social network topology corresponding to the public opinion data, comprising: A public opinion propagation topology graph is generated based on the social network topology corresponding to the public opinion data. A public opinion propagation search tree is constructed based on the public opinion propagation topology graph. 5.The public opinion data mining method of claim 3, wherein, The public opinion propagation search tree is generated based on the social network topology corresponding to the public opinion data, comprising: A public opinion propagation topology graph is generated based on the social network topology corresponding to the public opinion data. A public opinion propagation search tree is constructed based on the public opinion propagation topology graph.

6. The public opinion data mining method according to any one of claims 1 or 2 or 5, characterized in that, Before the public opinion data is converted into text data, if the public opinion data is video data, the voice signal corresponding to the video data is converted into text data; wherein the voice signal corresponding to the video data is converted into text data, comprising: determining the number of speakers of the voice signal; if the number of speakers is greater than 1, the identity corresponding to each speaker is determined; the speaking start time and speaking end time of each identity are determined; the audio signal corresponding to the speaking start time and speaking end time of each identity is converted into text data; otherwise, the voice signal is directly converted into text data; If the identity corresponding to the speaker is configured as a question identity and an answer identity, the text data corresponding to the question identity and the answer identity is processed to obtain text data conforming to a preset rule; wherein the text data corresponding to the question identity and the answer identity is processed to obtain text data conforming to a preset rule, comprising: extracting the first text feature and the second text feature corresponding to the first text data of the question identity and the second text data of the answer identity, respectively; calculating the similarity of the first text feature and the second text feature; based on the similarity and a preset similarity, the text data corresponding to the question identity and the answer identity is processed to obtain text data conforming to a preset rule. 7.The public opinion data mining method of claim 3, wherein, Before the converting the public opinion data into text data, further comprising: if the public opinion data is video data, converting a voice signal corresponding to the video data into text data; wherein the converting the voice signal corresponding to the video data into text data comprises: determining a number of speakers of the voice signal; if the number of speakers is greater than 1, respectively determining identities corresponding to the speakers; respectively determining a speech start time and a speech end time of each of the identities; converting an audio signal corresponding to the speech start time and the speech end time of each of the identities into text data; otherwise, directly converting the voice signal into text data; if the identities corresponding to the speakers are configured as a question identity and an answer identity; processing text data corresponding to the question identity and the answer identity to obtain text data meeting a preset rule; wherein the processing the text data corresponding to the question identity and the answer identity to obtain text data meeting a preset rule comprises: respectively extracting first text features and second text features corresponding to first text data of the question identity and second text data of the answer identity; calculating a similarity of the first text features and the second text features; based on the similarity and a preset similarity, processing the text data corresponding to the question identity and the answer identity to obtain text data meeting a preset rule. 8.The public opinion data mining method of claim 4, wherein, Before the converting the public opinion data into text data, further comprising: if the public opinion data is video data, converting a voice signal corresponding to the video data into text data; wherein the converting the voice signal corresponding to the video data into text data comprises: determining a number of speakers of the voice signal; if the number of speakers is greater than 1, respectively determining identities corresponding to the speakers; respectively determining a speech start time and a speech end time of each of the identities; converting an audio signal corresponding to the speech start time and the speech end time of each of the identities into text data; otherwise, directly converting the voice signal into text data; if the identities corresponding to the speakers are configured as a question identity and an answer identity; processing text data corresponding to the question identity and the answer identity to obtain text data meeting a preset rule; wherein the processing the text data corresponding to the question identity and the answer identity to obtain text data meeting a preset rule comprises: respectively extracting first text features and second text features corresponding to first text data of the question identity and second text data of the answer identity; calculating a similarity of the first text features and the second text features; based on the similarity and a preset similarity, processing the text data corresponding to the question identity and the answer identity to obtain text data meeting a preset rule. 9.The public opinion data mining method of claim 6, wherein, The processing of the text data corresponding to the question identity and the answer identity based on the similarity and a preset similarity to obtain text data meeting a preset rule comprises: if the similarity is greater than or equal to the preset similarity, the corresponding first text data and the second text data are configured as the text data meeting the preset rule.

10. The public opinion data mining method according to any one of claims 7 or 8, characterized in that, The processing of the text data corresponding to the question identity and the answer identity based on the similarity and a preset similarity to obtain text data meeting a preset rule comprises: if the similarity is greater than or equal to the preset similarity, the corresponding first text data and the second text data are configured as the text data meeting the preset rule. 11.A public opinion data mining method applied to a target user, comprising: The public opinion data mining method of any one of claims 1-10, comprising: recommending a target user based on the perception sensitivity of a plurality of users of the public opinion data under the same cognition or different cognitions.

12. A public opinion data mining method applied to a spokesperson, comprising: The public opinion data mining method of any one of claims 1-10, comprising: determining a target user based on the perception sensitivity of a plurality of users of the public opinion data under the same cognition or different cognitions; and recommending a spokesperson corresponding to the public opinion data based on the target user.

13. An opinion data mining system, comprising: Comprising: The analysis unit is used for obtaining public opinion data, including: obtaining at least one public opinion keyword corresponding to the to-be-mined public opinion data; mining the to-be-mined public opinion data according to the public opinion keyword to obtain public opinion data; generating a public opinion propagation search tree based on a social network topology corresponding to the public opinion data; calculating a propagation center value and a priori estimate value corresponding to a public opinion propagation search tree node by using the public opinion propagation search tree; tracing the public opinion data according to the propagation center value and the priori estimate value to determine a public opinion initial user corresponding to a public opinion propagation search tree source node of public opinion data propagation; determining a plurality of users corresponding to the public opinion data based on a social network topology corresponding to the public opinion initial user; converting the public opinion data into text data, and converting the text data into a word list; converting each word in the word list into a word vector by using a preset word vector model; converting the word vector into a public opinion analysis text vector; performing attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector; wherein the attitude analysis on the plurality of users corresponding to the public opinion data includes: obtaining emotional expression key information; performing attitude analysis on the plurality of users corresponding to the public opinion data based on emotional expression keywords or emotional expression stickers or images in the emotional expression key information; the conversion of the word vector into the public opinion analysis text vector includes: splicing the word vector to obtain the public opinion analysis text vector; the attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector includes: performing attitude classification on the plurality of users corresponding to the public opinion data by using the public opinion analysis text vector based on a preset multi-classification model to obtain a plurality of analysis results corresponding to the attitude analysis; the attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector further includes: obtaining a plurality of public opinion analysis text vectors corresponding to the public opinion data of the same user on a plurality of public opinion platforms or the same public opinion platform; splicing and fusing the plurality of public opinion analysis text vectors; performing attitude classification on each user by using the spliced and fused fusion public opinion analysis text vector based on a preset multi-classification model to obtain a plurality of analysis results corresponding to the attitude analysis of each user; The determination unit is used for determining a perception sensitivity corresponding to the user under the same cognition or different cognition based on the plurality of analysis results corresponding to the attitude analysis, including: determining a situation or a plot corresponding to the public opinion data; determining a plurality of mainstream values corresponding to the plurality of users based on the situation or the plot corresponding to the public opinion data; configuring a plurality of mainstream values ranked highest in turn in the plurality of mainstream values as cognitions corresponding to the plurality of users; classifying the plurality of users according to the cognitions corresponding to the plurality of users; determining a perception sensitivity corresponding to each classified user based on the classified user.

14. An opinion data mining system, comprising: Including: The analysis unit is used for obtaining public opinion data, including: obtaining at least one public opinion keyword corresponding to the to-be-mined public opinion data; mining the to-be-mined public opinion data according to the public opinion keyword to obtain public opinion data; generating a public opinion propagation search tree based on a social network topology corresponding to the public opinion data; calculating a propagation center value and a priori estimate value corresponding to a public opinion propagation search tree node by using the public opinion propagation search tree; tracing the public opinion data according to the propagation center value and the priori estimate value to determine a public opinion initial user corresponding to a public opinion propagation search tree source node of public opinion data propagation; determining a plurality of users corresponding to the public opinion data based on a social network topology corresponding to the public opinion initial user; converting the public opinion data into text data, and converting the text data into a word list; converting each word in the word list into a word vector by using a preset word vector model; converting the word vector into a public opinion analysis text vector; performing attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector; wherein the attitude analysis on the plurality of users corresponding to the public opinion data includes: obtaining emotional expression key information; performing attitude analysis on the plurality of users corresponding to the public opinion data based on emotional expression keywords or emotional expression stickers or images in the emotional expression key information; the conversion of the word vector into the public opinion analysis text vector includes: splicing the word vector to obtain the public opinion analysis text vector; the attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector includes: performing attitude classification on the plurality of users corresponding to the public opinion data by using the public opinion analysis text vector based on a preset multi-classification model to obtain a plurality of analysis results corresponding to the attitude analysis; the attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector further includes: obtaining a plurality of public opinion analysis text vectors corresponding to the public opinion data of the same user on a plurality of public opinion platforms or the same public opinion platform; splicing and fusing the plurality of public opinion analysis text vectors; performing attitude classification on each user by using the spliced and fused fusion public opinion analysis text vector based on a preset multi-classification model to obtain a plurality of analysis results corresponding to the attitude analysis of each user; and the determination unit is used for determining a perception sensitivity corresponding to the user under the same cognition or different cognitions based on the plurality of analysis results corresponding to the attitude analysis, including: determining a situation or a plot corresponding to the public opinion data; determining a plurality of mainstream values corresponding to the plurality of users based on the situation or the plot corresponding to the public opinion data; configuring a plurality of mainstream values ranked highest in turn in the plurality of mainstream values as cognitions corresponding to the plurality of users; classifying the plurality of users according to the cognitions corresponding to the plurality of users; and determining a perception sensitivity corresponding to each classified user based on the classified user. The first recommendation unit is configured to recommend a target user based on a perception sensitivity of a plurality of users corresponding to the public opinion data under the same cognition or different cognitions.

15. An opinion data mining system, comprising: The method comprises: The analysis unit is configured to obtain public opinion data, including: obtaining at least one public opinion keyword corresponding to the public opinion data to be mined; mining the public opinion data to be mined according to the public opinion keyword to obtain public opinion data; generating a public opinion propagation search tree based on a social network topology corresponding to the public opinion data; calculating a propagation center value and a priori estimate value corresponding to a public opinion propagation search tree node by using the public opinion propagation search tree; tracing the public opinion data according to the propagation center value and the priori estimate value to determine a public opinion initial user corresponding to a public opinion propagation search tree source node of the public opinion data propagation; determining a plurality of users corresponding to the public opinion data based on a social network topology corresponding to the public opinion initial user; converting the public opinion data into text data and converting the text data into a word list; converting each word in the word list into a word vector by using a preset word vector model; converting the word vector into a public opinion analysis text vector; performing attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector; wherein the attitude analysis on the plurality of users corresponding to the public opinion data includes: obtaining emotional expression key information; performing attitude analysis on the plurality of users corresponding to the public opinion data based on emotional expression keywords or emotional expression stickers or images in the emotional expression key information; the conversion of the word vector into the public opinion analysis text vector includes: splicing the word vector to obtain the public opinion analysis text vector; the attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector includes: performing attitude classification on the plurality of users corresponding to the public opinion data by using the public opinion analysis text vector based on a preset multi-classification model to obtain a plurality of analysis results corresponding to the attitude analysis; the attitude analysis on the plurality of users corresponding to the public opinion data based on the public opinion analysis text vector further includes: obtaining a plurality of public opinion analysis text vectors corresponding to the public opinion data of the same user on a plurality of public opinion platforms or the same public opinion platform; splicing and fusing the plurality of public opinion analysis text vectors; performing attitude classification on each user by using the spliced and fused fusion public opinion analysis text vector based on a preset multi-classification model to obtain a plurality of analysis results corresponding to the attitude analysis of each user. The determining unit is configured to determine the corresponding perception sensitivity of the user under the same cognition or different cognition based on the corresponding multiple analysis results of the attitude analysis, including: determining the corresponding situation or scenario of the public opinion data; determining the corresponding multiple mainstream values of the multiple users based on the corresponding situation or scenario of the public opinion data; configuring the multiple mainstream values with the highest ranking in turn as the cognition of the multiple users; classifying the multiple users according to the cognition of the multiple users; and determining the corresponding perception sensitivity of the classified users respectively based on the classified users. The second recommendation unit is configured to recommend the spokesperson corresponding to the public opinion data based on a target user.

16. An electronic device, comprising: The apparatus comprises: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to invoke the instructions stored in the memory to execute the public opinion data mining method according to any one of claims 1 to 10.

17. A computer-readable storage medium having stored thereon computer program instructions, wherein, The computer program instructions, when executed by the processor, implement the public opinion data mining method according to any one of claims 1 to 10. The computer program instructions, when executed by the processor, implement the public opinion data mining method according to any one of claims 1 to 10.

Images