Public opinion search method, search device, storage medium and terminal device

Abstract

The present invention discloses a public opinion search method, device, storage medium and terminal equipment. The public opinion search method includes: receiving a first search keyword input by a user, and searching for corresponding first public opinion data according to the first search keyword; performing cluster analysis on the searched first public opinion data according to a first preset classification method to obtain multiple information categories; calculating the first propagation heat of each information category, and arranging the information categories in descending order according to the first propagation heat; extracting the information categories that are located before a first preset quantity threshold from the descending information categories, and outputting the extracted information categories to the user. After searching for the corresponding public opinion data according to the search keyword, the present invention can divide the searched public opinion data into multiple information categories according to a preset classification method, and can return the corresponding public opinion data to the user according to the propagation heat of the information category, so as to facilitate the user to quickly find the public opinion information he wants and improve the search efficiency of the public opinion information.

Description

Technical Field

[0001] The present invention relates to the technical field of network public opinion processing, and in particular to a public opinion search method, a search device, a storage medium and a terminal device. Background Art

[0002] Public opinion is the abbreviation of public opinion. It refers to the social attitudes of the public as the subject towards the social managers, enterprises, individuals and other organizations as the object and their political, social and moral orientations around the occurrence, development and changes of intermediary social events in a certain social space. It is the sum of the beliefs, attitudes, opinions and emotions expressed by the masses on various phenomena and problems in society. Most of the current public opinion systems only extract and summarize the original public opinion data to understand the public's ideological dynamics, grasp the public's needs and grasp the direction of public opinion based on the extracted public opinion data. It is difficult for users to find a specific public opinion data accurately and quickly, and it is difficult to know the details of the public opinion event. Summary of the invention

[0003] The embodiments of the present invention provide a public opinion search method, a search device, a storage medium and a terminal device, which can search for public opinion data according to search keywords and can output the public opinion data in a classified manner to facilitate users to quickly find the public opinion information they want, thereby greatly improving the search efficiency of public opinion information.

[0004] A first aspect of an embodiment of the present invention provides a public opinion search method, comprising:

[0005] Receiving a first search keyword input by a user, and searching for corresponding first public opinion data according to the first search keyword;

[0006] Performing cluster analysis on the searched first public opinion data according to a first preset classification method to obtain multiple information categories;

[0007] Calculating the first diffusion heat of each major information category, and arranging the major information categories in descending order according to the first diffusion heat;

[0008] The information categories that are located before a first preset quantity threshold are extracted from the information categories arranged in descending order, and the extracted information categories are output to the user.

[0009] A second aspect of an embodiment of the present invention provides a public opinion search device, including:

[0010] A public opinion data search module, configured to receive a first search keyword input by a user, and search for corresponding first public opinion data according to the first search keyword;

[0011] A public opinion data clustering module, used to perform cluster analysis on the searched first public opinion data according to a first preset classification method to obtain multiple information categories;

[0012] A first diffusion heat calculation module, used to calculate the first diffusion heat of each information category, and arrange the information categories in descending order according to the first diffusion heat;

[0013] The information category extraction module is used to extract the information categories that are located before a first preset quantity threshold from the information categories arranged in descending order, and output the extracted information categories to the user.

[0014] According to a third aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the public opinion search method described in the first aspect are implemented.

[0015] According to a fourth aspect of an embodiment of the present invention, a terminal device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer-readable instructions:

[0016] Receiving a first search keyword input by a user, and searching for corresponding first public opinion data according to the first search keyword;

[0017] Performing cluster analysis on the searched first public opinion data according to a first preset classification method to obtain multiple information categories;

[0018] Calculating the first diffusion heat of each major information category, and arranging the major information categories in descending order according to the first diffusion heat;

[0019] The information categories that are located before a first preset quantity threshold are extracted from the information categories arranged in descending order, and the extracted information categories are output to the user.

[0020] It can be seen from the above technical solutions that the embodiments of the present invention have the following advantages:

[0021] In an embodiment of the present invention, a search keyword input by a user can be received in a preset public opinion system, and corresponding public opinion data can be searched according to the search keyword. After the corresponding public opinion data is searched, the searched public opinion data can be divided into multiple information categories according to a preset classification method, and the dissemination heat of each information category can be counted. Subsequently, multiple information categories with the highest dissemination heat are extracted and returned to the user, thereby achieving the purpose of searching public opinion data according to the search keyword and outputting the public opinion data according to the dissemination heat classification, so as to facilitate users to quickly find the public opinion information they want, thereby greatly improving the search efficiency of public opinion information. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0023] Figure 1 A flowchart of a method for searching public opinion in an embodiment of the present invention;

[0024] Figure 2 A schematic diagram of a process of arranging information subcategories in an application scenario according to a public opinion search method in an embodiment of the present invention;

[0025] Figure 3 A schematic diagram of a process of determining a major information category corresponding to a second search keyword in an application scenario according to a public opinion search method in an embodiment of the present invention;

[0026] Figure 4 This is a structural diagram of an embodiment of a public opinion search device in an embodiment of the present invention;

[0027] Figure 5 A schematic diagram of a terminal device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The embodiments of the present invention provide a public opinion search method, a search device, a storage medium and a terminal device, which are used to search public opinion data according to search keywords and output the public opinion data in a classified manner, so as to facilitate users to quickly find the public opinion information they want, thereby greatly improving the search efficiency of public opinion information.

[0029] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In addition, the terms "first", "second" and "third" are used to distinguish different objects, rather than to describe a specific order.

[0030] In an embodiment of the present invention, the executor of the public opinion search method is a preset public opinion system, and when executing the public opinion search method, the preset public opinion system has crawled public opinion data related to each public opinion event through crawling technologies such as a distributed crawler system, and has stored the crawled public opinion data in a database of the preset public opinion system, such as storing the crawled public opinion data in a storage server corresponding to the preset public opinion system. Therefore, when the preset public opinion system receives a search keyword, the preset public opinion system can directly search for public opinion data matching the search keyword in the corresponding storage server.

[0031] See also Figure 1 , an embodiment of the present invention provides a public opinion search method, the public opinion search method comprising:

[0032] Step S101: receiving a first search keyword input by a user, and searching for corresponding first public opinion data according to the first search keyword;

[0033] It can be understood that a public opinion retrieval portal is provided in the preset public opinion system of the embodiment of the present invention. When a user needs to search for a public opinion event, the user can input the search keyword corresponding to the public opinion event through the public opinion retrieval portal. For example, the search keywords of "football" and "transfer" can be input through the public opinion retrieval portal. After receiving the search keywords of "football" and "transfer" input by the user, the preset public opinion system can search for the first public opinion data related to "football" and "transfer" in the database of the preset public opinion system. That is to say, the first public opinion data related to transfer in the football field is searched in the storage server corresponding to the preset public opinion system.

[0034] Step S102: performing cluster analysis on the searched first public opinion data according to a first preset classification method to obtain multiple information categories;

[0035] Here, after searching and obtaining the corresponding first public opinion data, the preset public opinion system can perform cluster analysis on the searched first public opinion data according to the first preset classification method to obtain multiple information categories, such as performing cluster analysis on the searched first public opinion data according to the public opinion subjects involved in the searched first public opinion data, so as to classify the first public opinion data of the same subject into the same information category. For example, when the public opinion subjects involved in the searched first public opinion data include star A, star B and star C, the first public opinion data related to the transfer of star A can be classified into information category A, the first public opinion data related to the transfer of star B can be classified into information category B, and the first public opinion data related to the transfer of star C can be classified into information category C.

[0036] Among them, in an embodiment of the present invention, the first preset classification method can be specifically determined according to the actual situation of the public opinion event searched by the user. For example, when the public opinion event searched by the user involves different public opinion subjects, the first preset classification method may be a method of classification according to the public opinion subject; for another example, when the public opinion event searched by the user involves different communication channels or communication sources, the first preset classification method may be a method of classification according to the communication channels and / or communication sources; for another example, when the public opinion event searched by the user involves different time periods, the first preset classification method may be a method of classification according to time periods, and so on.

[0037] Step S103, calculating the first diffusion heat of each information category, and arranging the information categories in descending order according to the first diffusion heat;

[0038] In the embodiment of the present invention, after clustering analysis is performed on the searched first public opinion data to obtain multiple information categories, the first transmission heat of each information category can be further calculated, and each information category can be arranged in descending order according to the first transmission heat, that is, the information category with a larger first transmission heat will be ranked higher, and the information category with a smaller first transmission heat will be ranked lower. The calculation of the first transmission heat of each information category will be described in detail in the subsequent content.

[0039] For example, in a specific application, when the information categories obtained by cluster analysis of the searched first public opinion data include the first information category, the second information category, the third information category and the fourth information category, and it is calculated that the first propagation heat of the first information category is 500, the first propagation heat of the second information category is 800, the first propagation heat of the third information category is 4000, and the first propagation heat of the fourth information category is 2700, then the information categories can be arranged in descending order to obtain a list of information categories (third information category, fourth information category, second information category, first information category).

[0040] Step S104: extracting the information categories before a first preset quantity threshold from the information categories arranged in descending order, and outputting the extracted information categories to the user.

[0041] It should be noted that, in an embodiment of the present invention, when extracting data located before a preset quantity threshold from data arranged in descending order, the extracted data includes the data within the preset quantity threshold. For example, when a preset quantity threshold is determined to be 5, the data extracted from the data located before 5 in the descending order may include the data ranked first to fifth.

[0042] It can be understood that in the embodiment of the present invention, the preset public opinion system can automatically determine the number of major information categories to be output to the user based on the public opinion event, and can also set the default number of major information categories to be output to the user when constructing the preset public opinion system. Of course, the user can also set the number of major information categories he wants by himself. Therefore, when the user sets the number of major information categories he wants by himself, the number set by the user can be determined as the first preset number threshold. When the user does not set the number of major information categories he wants by himself, the preset public opinion system can automatically determine the number determined by the public opinion event as the first preset number threshold, or the default output number set when constructing the preset public opinion system can be determined as the first preset number threshold, so as to extract the major information categories located before the first preset number threshold from the list of major information categories arranged in descending order and output them to the user, that is, extract the major information categories with the first preset number threshold number ranked in front from the list of major information categories arranged in descending order and output them to the user.

[0043] For example, in a specific application, when the user does not set the number of information categories he wants, and the preset public opinion system automatically determines that the number of information categories to be output to the user is 5 based on the public opinion event, and the list of information categories obtained in descending order is (third information category, fourth information category, second information category, first information category, seventh information category, ninth information category, sixth information category, eighth information category, fifth information category), then the top 5 information categories in the list of information categories can be extracted and output to the user, that is, the third information category, fourth information category, second information category, first information category and seventh information category can be output to the user.

[0044] Furthermore, if Figure 2 As shown, in an embodiment of the present invention, after clustering analysis is performed on the searched first public opinion data according to the first preset classification method to obtain multiple information categories, the following may also be included:

[0045] Step S201: performing cluster analysis on the second public opinion data in the information category according to a second preset classification method to obtain multiple information subcategories;

[0046] Step S202, calculating the second propagation heat of each information subcategory;

[0047] Step S203: Arrange the information subcategories in descending order in the information major category according to the second propagation heat.

[0048] For the above steps S201 to S203, it can be understood that after clustering the first public opinion data to obtain multiple information categories, further clustering analysis can be performed on the second public opinion data in each information category according to the second preset classification method to obtain multiple information subcategories in each information category. The second public opinion data refers to the first public opinion data clustered into each information category according to the first preset classification method.

[0049] Here, the second preset classification method can be a method of classifying according to the subject of public opinion, a method of classifying according to the communication channel or source of communication, a method of classifying according to the time period, etc. It can be understood that in the search of the same public opinion event, the second preset classification method should be a classification method different from the first preset classification method.

[0050] For example, in the specific application scenario of star transfers, after clustering the first public opinion data according to the public opinion subject to obtain information category A related to the transfer of star A, information category B related to the transfer of star B, and information category C related to the transfer of star C, the public opinion data in information category A, information category B, and information category C can be further clustered according to the communication channels, so as to further divide the public opinion data in each information category into multiple information subcategories based on the communication channels. For example, the public opinion data about the transfer of star A in information category A is divided into a first information subcategory, a second information subcategory, and a third information subcategory, wherein the first information subcategory is the public opinion data related to the transfer of star A from social media, the second information subcategory is the public opinion data related to the transfer of star A from news media, and the third information subcategory is the public opinion data related to the transfer of star A from forums. For the public opinion data about the transfer of star B in information category B and the public opinion data about the transfer of star C in information category C, the same division can be performed based on the communication channels to obtain corresponding multiple information subcategories in information category B and information category C.

[0051] Furthermore, after obtaining each information subcategory in a certain information category, the second transmission heat of each information subcategory in the information category can be further calculated, and then each information subcategory in the information category can be arranged in descending order according to the second transmission heat, that is, the information subcategory with a larger second transmission heat will be ranked higher, and the information subcategory with a smaller second transmission heat will be ranked lower, so that when the user clicks to enter a certain information category, each information subcategory can be displayed to the user according to the transmission heat, so that the user can quickly find the key information and critical information of the public opinion event, so as to facilitate the user to quickly understand the development of the public opinion event. Among them, the calculation of the second transmission heat of each information subcategory will also be described in detail in the subsequent content.

[0052] Preferably, after extracting the information categories before the first preset quantity threshold from the information categories arranged in descending order and outputting the extracted information categories to the user, the method may include:

[0053] Step a, receiving a second search keyword input by a user;

[0054] Step b, determining the information category corresponding to the second search keyword;

[0055] Step c: searching for third public opinion data matching the second search keyword from the determined information categories, and outputting the searched third public opinion data to the user.

[0056] For the above steps a to c, it can be understood that after the user obtains the search results output by the preset public opinion system for the first time, the user can also conduct a secondary search in the search results to accurately find the public opinion data he wants, that is, after the preset public opinion system returns the corresponding information category in the first search, the user can further enter a second search keyword. After receiving the second search keyword, the preset public opinion system can first determine the information category corresponding to the second search keyword, and then further search for the third public opinion data matching the second search keyword in the determined information category. For example, in the specific application scenario of star transfers, after obtaining information category A, information category B, information category C and information category D returned by the preset public opinion system in the first search, the user can further enter the second search keyword "membership fee for star A's transfer". The preset public opinion system can determine the corresponding information category based on "membership fee for star A's transfer". For example, when it is determined that the information category corresponding to "membership fee for star A's transfer" is information category A, the preset public opinion system can further search for the third public opinion data matching the "membership fee for star A's transfer" in information category A, and output the searched third public opinion data to the user.

[0057] It should be noted that, in an embodiment of the present invention, after receiving the first search keyword and the second search keyword, the preset public opinion system may also first perform pre-processing such as cleaning and extraction on the first search keyword and the second search keyword to remove invalid words, thereby ensuring the conciseness and effectiveness of the search keywords, thereby obtaining more accurate search results.

[0058] Alternatively, if Figure 3 As shown, in the embodiment of the present invention, the determining of the information category corresponding to the second search keyword may include:

[0059] Step S301, obtaining core keywords corresponding to each major information category, wherein the core keywords refer to keywords extracted from the second public opinion data of each major information category by a preset extraction method;

[0060] It can be understood that in the embodiment of the present invention, after clustering analysis is performed on the searched first public opinion data to obtain multiple information categories, the second public opinion data in each information category can be first extracted, and the second public opinion data is a part of the first public opinion data; then the extracted second public opinion data is subjected to text training to obtain a word vector model and a word frequency-inverse document frequency TF_IDF matrix, and the first keyword corresponding to each information category is determined through the TF_IDF matrix; then, each first keyword can be expanded through the word vector model, that is, each first keyword is input into the word vector model respectively, so as to obtain similar words with semantic similarity to each first keyword output by the word vector model and the corresponding semantic similarity, and similar words with semantic similarity reaching a preset similarity threshold are determined as second keywords related to the corresponding first keyword, and the first keyword and the related second keyword are determined as core keywords corresponding to the corresponding information category. Among them, the preset similarity threshold can be specifically set or automatically determined according to the actual application scenario.

[0061] For example, in a specific application scenario, when the first keyword corresponding to information category A is determined to be [a,b,c] according to the TF_IDF matrix, and the second keyword obtained after expanding the first keyword according to the word vector model is [a1,a2,b1,b2,b3], the first keyword [a,b,c] and the second keyword [a1,a2,b1,b2,b3] can be determined together as the core keywords corresponding to information category A, that is, the core keywords corresponding to information category A are [a,b,c,a1,a2,b1,b2,b3], wherein a1 and a2 are words determined by the word vector model and whose semantic similarity with a reaches the preset similarity threshold, and b1, b2 and b3 are words determined by the word vector model and whose semantic similarity with b reaches the preset similarity threshold.

[0062] Step S302: vectorize the core keyword and the second search keyword respectively to obtain a first vector corresponding to the core keyword and a second vector corresponding to the second search keyword;

[0063] Here, after determining the core keywords corresponding to each information category, each core keyword can be vectorized to obtain a first vector corresponding to each core keyword, and the received second search keyword can also be vectorized to obtain a second vector corresponding to the second search keyword. Here, when there is more than one second search keyword, each second search keyword can be vectorized to obtain multiple initial vectors, and then the average vector of the multiple initial vectors can be calculated, and the obtained average vector is determined as the second vector corresponding to the second search keyword.

[0064] Step S303, calculating the cosine similarity between the second vector and each of the first vectors;

[0065] Step S304: obtaining a first vector whose cosine similarity is greater than a preset similarity threshold, and determining the information category corresponding to the obtained first vector as the information category corresponding to the second search keyword.

[0066] For the above steps S303 and S304, it can be understood that after obtaining the first vector corresponding to each core keyword and the second vector corresponding to the second search keyword, the cosine similarity formula can be used to calculate the cosine similarity between the second vector and each first vector, so as to determine the core keyword matching the second search keyword according to the cosine similarity. For example, when the cosine similarity between the second vector and the first vector corresponding to the core keyword r is greater than the preset cosine threshold, it can be determined that the second search keyword matches the core keyword r, and thus, the information category corresponding to the core keyword r can be determined as the information category corresponding to the second search keyword. Here, the preset cosine threshold can be specifically determined according to the actual application scenario.

[0067] It should be noted that, in the embodiment of the present invention, of course, the cosine similarity formula can also be used to calculate the cosine similarity between the second vector and each first vector respectively, and then the first vector corresponding to the maximum cosine similarity is obtained, and the core keyword corresponding to the obtained first vector is determined as the core keyword matching the second search keyword. For example, when the cosine similarity between the second vector and the first vector corresponding to the core keyword q is 0.8, the cosine similarity between the second vector and the first vector corresponding to the core keyword s is 0.3, and the cosine similarity between the second vector and the first vector corresponding to the core keyword t is 0.4, it can be determined that the second search keyword matches the core keyword q, that is, the information category corresponding to the core keyword q is the information category corresponding to the second search keyword.

[0068] Further, in the embodiment of the present invention, searching the determined information categories for third public opinion data matching the second search keyword and outputting the searched third public opinion data to the user may include:

[0069] Step d: searching for third public opinion data matching the second search keyword from each information subcategory of the determined information category;

[0070] Step e: calculating the third propagation heat of the searched third public opinion data, and arranging the third public opinion data in descending order according to the third propagation heat;

[0071] Step f: extracting the fourth public opinion data that is located before a second preset quantity threshold from the third public opinion data arranged in descending order, and outputting the extracted fourth public opinion data to the user.

[0072] For the above steps d to f, it can be understood that, since the corresponding public opinion data has been divided into multiple information subcategories in each information category, after determining the information category corresponding to the second search keyword, the public opinion data matching the second search keyword can be searched in each information subcategory of the information category, that is, the third public opinion data mentioned above, wherein the third public opinion data is part or all of the second public opinion data; after obtaining the third public opinion data, the propagation heat of the third public opinion data can be further calculated, that is, the third propagation heat mentioned above, and each third public opinion data can be arranged in descending order according to the third propagation heat, that is, the third public opinion data with a larger third propagation heat will be ranked higher, and the third public opinion data with a smaller third propagation heat will be ranked lower; after descending order, the public opinion data with a second preset number threshold number that is ranked first can be extracted and output to the user, that is, one or more public opinion data with a large propagation heat can be output to the user. Among them, the second preset number threshold can be specifically determined according to the actual application scenario.

[0073] Preferably, in the embodiment of the present invention, the calculating the second propagation heat of each information subcategory may include:

[0074] Obtain the reading volume, comment volume, and forwarding volume corresponding to the public opinion data in each information subcategory, and calculate the fourth communication heat of the public opinion data in each information subcategory according to the following formula:

[0075] WeightDate t,m,i =a t,m,i *Read t,m,i +b t,m,i *Comments t,m,i +c t,m,i *Repost t,m,i ;

[0076] Among them, WeightDate t,m,i is the fourth diffusion heat of the ith public opinion data in the mth information subcategory of the tth information category, Read t,m,i is the reading volume of the i-th public opinion data, a t,m,i is the reading weight of the i-th public opinion data, Comments t,m,i is the number of comments on the i-th public opinion data, b t,m,i is the comment weight of the i-th public opinion data, Repost t,m,i is the forwarding amount of the i-th public opinion data, c t,m,i is the forwarding weight of the i-th public opinion data;

[0077] The second transmission heat of each information subcategory is calculated according to the following formula:

[0078]

[0079] Among them, WeightClassS t,m is the second diffusion heat of the mth information subcategory in the tth information category, n is the number of public opinion data in the mth information subcategory, WeightDate t,m,i It is the fourth dissemination heat of the ith public opinion data in the mth information subcategory of the tth information category.

[0080] Optionally, calculating the first propagation heat of each major information category may include:

[0081] The first dissemination heat of each information category is calculated according to the following formula:

[0082]

[0083] Among them, WeightClassB t is the first diffusion heat of the t-th information category, p is the number of information subcategories in the t-th information category, WeightClassS t,m It is the second diffusion heat of the mth information category in the tth information category.

[0084] It is understandable that in the embodiment of the present invention, the spread heat of each public opinion data is evaluated mainly based on the reading volume, forwarding volume and comment volume of the public opinion data, that is, the fourth spread heat mentioned above, that is, the public opinion data with larger reading volume, comment volume and forwarding volume will have higher spread heat, and the public opinion data with smaller reading volume, comment volume and forwarding volume will have lower spread heat. Preferably, in the embodiment of the present invention, when evaluating the spread heat of public opinion data, the release time of the public opinion data can also be considered. For example, the public opinion data with newer release time can have corresponding increase in spread heat, while the public opinion data with older release time can have corresponding decrease in spread heat.

[0085] Furthermore, in an embodiment of the present invention, the reading weight, comment weight and forwarding weight of each public opinion data can be set according to the actual application scenario, wherein the reading weight of each public opinion data can be the same or different; the comment weight of each public opinion data can be the same or different; similarly, the forwarding weight of each public opinion data can be the same or different.

[0086] Here, after obtaining the fourth propagation heat of each public opinion data, the fourth propagation heat of all public opinion data belonging to the same information subcategory can be added to obtain the propagation heat of each information subcategory, that is, the second propagation heat mentioned above. After obtaining the second propagation heat of each information subcategory, the second propagation heat of all information subcategories in the same information major category can be added to obtain the propagation heat of each information major category, that is, the first propagation heat mentioned above.

[0087] In an embodiment of the present invention, a search keyword input by a user can be received in a preset public opinion system, and corresponding public opinion data can be searched according to the search keyword. After the corresponding public opinion data is searched, the searched public opinion data can be divided into multiple information categories according to a preset classification method, and the dissemination heat of each information category can be counted. Subsequently, multiple information categories with the highest dissemination heat are extracted and returned to the user, thereby achieving the purpose of searching public opinion data according to the search keyword and outputting the public opinion data according to the dissemination heat classification, so as to facilitate users to quickly find the public opinion information they want, thereby greatly improving the search efficiency of public opinion information.

[0088] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0089] The above mainly describes a public opinion search method, and the following will describe a public opinion search device in detail.

[0090] like Figure 4 As shown, an embodiment of the present invention provides a public opinion search device, the public opinion search device comprising:

[0091] The public opinion data search module 401 is used to receive a first search keyword input by a user, and search for corresponding first public opinion data according to the first search keyword;

[0092] The public opinion data clustering module 402 is used to perform cluster analysis on the searched first public opinion data according to a first preset classification method to obtain multiple information categories;

[0093] A first diffusion heat calculation module 403 is used to calculate the first diffusion heat of each information category and arrange the information categories in descending order according to the first diffusion heat;

[0094] The information category extraction module 404 is used to extract the information categories that are located before the first preset quantity threshold from the information categories arranged in descending order, and output the extracted information categories to the user.

[0095] Furthermore, the public opinion search device further includes:

[0096] An information subcategory acquisition module, used to perform cluster analysis on the second public opinion data in the information major category according to a second preset classification method to obtain multiple information subcategories;

[0097] The second propagation heat calculation module is used to calculate the second propagation heat of each information subcategory;

[0098] The module for arranging the information subcategories in descending order is used to arrange the information subcategories in descending order in the information major category according to the second propagation heat.

[0099] Preferably, the public opinion search device comprises:

[0100] A second search keyword receiving module, used to receive a second search keyword input by a user;

[0101] An information category determination module, used to determine the information category corresponding to the second search keyword;

[0102] The second search keyword search module is used to search for third public opinion data matching the second search keyword from the determined information category, and output the searched third public opinion data to the user.

[0103] Optionally, the information category determination module includes:

[0104] A core keyword acquisition unit, used to acquire core keywords corresponding to each information category, wherein the core keywords refer to keywords extracted from the second public opinion data of each information category by a preset extraction method;

[0105] A vectorization processing unit, used to perform vectorization processing on the core keyword and the second search keyword respectively to obtain a first vector corresponding to the core keyword and a second vector corresponding to the second search keyword;

[0106] a cosine similarity calculation unit, configured to calculate the cosine similarity between the second vector and each of the first vectors;

[0107] The information category determination unit is used to obtain a first vector whose cosine similarity is greater than a preset similarity threshold, and determine the information category corresponding to the obtained first vector as the information category corresponding to the second search keyword.

[0108] Furthermore, the second search keyword search module includes:

[0109] A second search keyword search unit, used to search for third public opinion data matching the second search keyword from each information subcategory of the determined information category;

[0110] A third propagation heat calculation unit, used to calculate the third propagation heat of the searched third public opinion data, and to arrange the third public opinion data in descending order according to the third propagation heat;

[0111] The public opinion data descending order arrangement unit is used to extract the fourth public opinion data located before the second preset quantity threshold from the third public opinion data arranged in descending order, and output the extracted fourth public opinion data to the user.

[0112] Preferably, the second propagation heat calculation module includes:

[0113] The first calculation unit is used to obtain the reading volume, comment volume and forwarding volume corresponding to the public opinion data in each information subcategory, and calculate the fourth communication heat of the public opinion data in each information subcategory according to the following formula: WeightDate t,m,i =a t,m,i *Read t,m,i +b t,m,i *Comments t,m,i +c t,m,i *Repost t,m,i ;

[0114] Among them, WeightDate t,m,i is the fourth diffusion heat of the ith public opinion data in the mth information subcategory of the tth information category, Read t,m,i is the reading volume of the i-th public opinion data, a t,m,i is the reading weight of the i-th public opinion data, Comments t,m,i is the number of comments on the i-th public opinion data, bt,m,i is the comment weight of the i-th public opinion data, Repost t,m,i is the forwarding amount of the i-th public opinion data, c t,m,i is the forwarding weight of the i-th public opinion data;

[0115] The second calculation unit is used to calculate the second propagation heat of each information subcategory according to the following formula:

[0116]

[0117] Among them, WeightClassS t,m is the second diffusion heat of the mth information subcategory in the tth information category, n is the number of public opinion data in the mth information subcategory, WeightDate t,m,i It is the fourth dissemination heat of the ith public opinion data in the mth information subcategory of the tth information category.

[0118] Optionally, the first propagation heat calculation module 403 is used to calculate the first propagation heat of each information category according to the following formula:

[0119]

[0120] Among them, WeightClassB t is the first diffusion heat of the t-th information category, p is the number of information subcategories in the t-th information category, WeightClassS t,m It is the second diffusion heat of the mth information category in the tth information category.

[0121] Figure 5 Schematic diagram of a terminal device provided by an embodiment of the present invention. Figure 5 As shown, the terminal device 5 of this embodiment includes: a processor 50, a memory 51, and a computer-readable instruction 52 stored in the memory 51 and executable on the processor 50, such as a public opinion search program. When the processor 50 executes the computer-readable instruction 52, the steps in the above-mentioned public opinion search method embodiments are implemented, such as Figure 1 Alternatively, when the processor 50 executes the computer-readable instructions 52, the functions of each module / unit in the above-mentioned device embodiments are realized, for example, Figure 4 Functions of modules 401 to 404 are shown.

[0122] Exemplarily, the computer-readable instructions 52 may be divided into one or more modules / units, which are stored in the memory 51 and executed by the processor 50 to implement the present invention. The one or more modules / units may be a series of computer-readable instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer-readable instructions 52 in the terminal device 5.

[0123] The terminal device 5 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor 50 and a memory 51. Those skilled in the art will appreciate that Figure 5 It is only an example of the terminal device 5 and does not constitute a limitation of the terminal device 5. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.

[0124] The processor 50 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0125] The memory 51 may be an internal storage unit of the terminal device 5, such as a hard disk or memory of the terminal device 5. The memory 51 may also be an external storage device of the terminal device 5, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 5. Further, the memory 51 may also include both an internal storage unit of the terminal device 5 and an external storage device. The memory 51 is used to store the computer-readable instructions and other programs and data required by the terminal device. The memory 51 may also be used to temporarily store data that has been output or is to be output.

[0126] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0127] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0128] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for searching public opinion, characterized in that: include: Receiving a first search keyword input by a user, and searching for corresponding first public opinion data according to the first search keyword; Performing cluster analysis on the searched first public opinion data according to a first preset classification method to obtain multiple information categories; Calculating the first diffusion heat of each major information category, and arranging the major information categories in descending order according to the first diffusion heat; Extracting the information categories before a first preset number threshold from the information categories arranged in descending order, and outputting the extracted information categories to the user; Wherein, after clustering the searched first public opinion data according to the first preset classification method to obtain multiple information categories, the method further includes: Performing cluster analysis on the second public opinion data in the information category according to a second preset classification method to obtain a plurality of information subcategories in each information category, wherein the second preset classification method is different from the first preset classification method; Calculate the second transmission heat of each information subcategory; Arrange the information subcategories in descending order in the information major category according to the second propagation heat; The calculating the second propagation heat of each information subcategory includes: Obtain the reading volume, comment volume, and forwarding volume corresponding to the public opinion data in each information subcategory, and calculate the fourth communication heat of the public opinion data in each information subcategory according to the following formula: = ； in, is the fourth diffusion heat of the ith public opinion data in the mth information subcategory of the tth information category, is the reading volume of the i-th public opinion data, is the reading weight of the i-th public opinion data, is the number of comments on the i-th public opinion data, is the comment weight of the i-th public opinion data, is the forwarding amount of the i-th public opinion data, is the forwarding weight of the i-th public opinion data; The second transmission heat of each information subcategory is calculated according to the following formula: = ； in, is the second diffusion heat of the mth information subcategory in the tth information category, n is the number of public opinion data in the mth information subcategory, It is the fourth dissemination heat of the ith public opinion data in the mth information subcategory of the tth information category.

2. The public opinion search method according to claim 1, characterized in that: After extracting the information categories before the first preset quantity threshold from the information categories arranged in descending order and outputting the extracted information categories to the user, the method includes: receiving a second search keyword input by a user; Determining a major category of information corresponding to the second search keyword; The third public opinion data matching the second search keyword is searched from the determined information categories, and the searched third public opinion data is output to the user.

3. The public opinion search method according to claim 2, characterized in that: The determining of the information category corresponding to the second search keyword includes: Obtaining core keywords corresponding to each major information category, wherein the core keywords refer to keywords extracted from the second public opinion data of each major information category by a preset extraction method; Performing vectorization processing on the core keyword and the second search keyword respectively to obtain a first vector corresponding to the core keyword and a second vector corresponding to the second search keyword; Calculating the cosine similarity between the second vector and each of the first vectors; A first vector whose cosine similarity is greater than a preset similarity threshold is obtained, and the information category corresponding to the obtained first vector is determined as the information category corresponding to the second search keyword.

4. The public opinion search method according to claim 2, characterized in that: The step of searching the determined information categories for third public opinion data that matches the second search keyword, and outputting the searched third public opinion data to the user includes: Searching for third public opinion data matching the second search keyword from each information subcategory of the determined information category; Calculating the third propagation heat of the searched third public opinion data, and arranging the third public opinion data in descending order according to the third propagation heat; The fourth public opinion data located before a second preset quantity threshold is extracted from the third public opinion data arranged in descending order, and the extracted fourth public opinion data is output to the user.

5. The public opinion search method according to any one of claims 1 to 4, characterized in that: The calculating of the first propagation heat of each information category includes: The first dissemination heat of each information category is calculated according to the following formula: = ； in, is the first diffusion heat of the t-th information category, p is the number of information subcategories in the t-th information category, It is the second diffusion heat of the mth information category in the tth information category.

6. A public opinion search device, characterized in that: include: A public opinion data search module, configured to receive a first search keyword input by a user, and search for corresponding first public opinion data according to the first search keyword; A public opinion data clustering module, used to perform cluster analysis on the searched first public opinion data according to a first preset classification method to obtain multiple information categories; A first diffusion heat calculation module, used to calculate the first diffusion heat of each information category, and arrange the information categories in descending order according to the first diffusion heat; An information category extraction module, used to extract information categories before a first preset number threshold from the information categories arranged in descending order, and output the extracted information categories to the user; An information subcategory acquisition module, used for performing cluster analysis on the second public opinion data in the information major category according to a second preset classification method to obtain a plurality of information subcategories in each information major category, wherein the second preset classification method is different from the first preset classification method; The second propagation heat calculation module is used to calculate the second propagation heat of each information subcategory; An information subcategory descending order arrangement module, used for arranging the information subcategories in descending order in the information major category according to the second propagation heat; The second propagation heat calculation module is specifically used for: Obtain the reading volume, comment volume, and forwarding volume corresponding to the public opinion data in each information subcategory, and calculate the fourth communication heat of the public opinion data in each information subcategory according to the following formula: = ； in, is the fourth diffusion heat of the ith public opinion data in the mth information subcategory of the tth information category, is the reading volume of the i-th public opinion data, is the reading weight of the i-th public opinion data, is the number of comments on the i-th public opinion data, is the comment weight of the i-th public opinion data, is the forwarding amount of the i-th public opinion data, is the forwarding weight of the i-th public opinion data; The second transmission heat of each information subcategory is calculated according to the following formula: = ； in, is the second diffusion heat of the mth information subcategory in the tth information category, n is the number of public opinion data in the mth information subcategory, It is the fourth dissemination heat of the ith public opinion data in the mth information subcategory of the tth information category.

7. A computer-readable storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by a processor, the steps of the public opinion search method as described in any one of claims 1 to 5 are implemented.

8. A terminal device, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the following steps are implemented: Receiving a first search keyword input by a user, and searching for corresponding first public opinion data according to the first search keyword; Performing cluster analysis on the searched first public opinion data according to a first preset classification method to obtain multiple information categories; Calculating the first diffusion heat of each major information category, and arranging the major information categories in descending order according to the first diffusion heat; Extracting the information categories before a first preset number threshold from the information categories arranged in descending order, and outputting the extracted information categories to the user; Wherein, after clustering the searched first public opinion data according to the first preset classification method to obtain multiple information categories, the method further includes: Performing cluster analysis on the second public opinion data in the information category according to a second preset classification method to obtain a plurality of information subcategories in each information category, wherein the second preset classification method is different from the first preset classification method; Calculate the second transmission heat of each information subcategory; Arrange the information subcategories in descending order in the information major category according to the second propagation heat; The calculating the second propagation heat of each information subcategory includes: Obtain the reading volume, comment volume, and forwarding volume corresponding to the public opinion data in each information subcategory, and calculate the fourth communication heat of the public opinion data in each information subcategory according to the following formula: = ； in, is the fourth diffusion heat of the ith public opinion data in the mth information subcategory of the tth information category, is the reading volume of the i-th public opinion data, is the reading weight of the i-th public opinion data, is the number of comments on the i-th public opinion data, is the comment weight of the i-th public opinion data, is the forwarding amount of the i-th public opinion data, is the forwarding weight of the i-th public opinion data; The second transmission heat of each information subcategory is calculated according to the following formula: = ； in, is the second diffusion heat of the mth information subcategory in the tth information category, n is the number of public opinion data in the mth information subcategory, It is the fourth dissemination heat of the ith public opinion data in the mth information subcategory of the tth information category.

Images