Comment information processing method and device, electronic equipment and storage medium
By analyzing the relevance, positive and negative aspects, and sentiment of comment information, combined with clustering algorithms and screening strategies, the problem of users having difficulty quickly obtaining comment content is solved, the interactivity and user experience of the video comment system are improved, and false information is reduced.
Patent Information
- Application Number
- CN202410410747.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-07
- Publication Date
- 2025-10-17
AI Technical Summary
In existing video comment systems, it is difficult for users to quickly obtain large amounts of comment content and express their opinions. The display of comment content is not intuitive enough, and there is a lack of effective screening and search functions. Anonymous comments lead to a decline in quality and an increased possibility of false information spreading.
By conducting correlation analysis, positive and negative analysis, and sentiment analysis on comment information, adopting a multi-dimensional analysis method, combined with clustering algorithms and screening strategies, we can extract user-interest-related, real-time, and customized comment content, and provide an intelligent reading mode.
It enables users to quickly obtain key information from comments, improves interactivity and fun, enhances user experience, reduces false information, and enhances the quality of the comment area.
Smart Images
Figure CN120805887A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present disclosure relates to the technical field of computer, in particular to the technical field of data processing, and more particularly to a comment information processing method and device, an electronic device and a storage medium. BACKGROUND
[0002] With the rapid development of videos, short video comments are an integral part of social media platforms, which promote interaction, feedback, information dissemination and other multiple effects, establish a closer community relationship between users and creators, and enrich the user experience of social media.
[0003] Most video software on the market displays video comments in the form of a stream, and the comment content is generally divided into two levels, replies and other forms. Reading and browsing requires users to manually slide to view and read, and two-level comments need to be manually expanded and closed. When browsing too many comment contents, such as tens of thousands of video comment contents or more, users cannot quickly obtain all the comment contents and express their opinions. SUMMARY
[0004] The present disclosure provides a comment information processing method and device, an electronic device and a storage medium.
[0005] According to an aspect of the present disclosure, a comment information processing method is provided, comprising:
[0006] obtaining comment information, the comment information comprising a comment object, a comment content and video content corresponding to the comment content;
[0007] performing correlation analysis, positive and negative analysis and sentiment analysis on the comment information respectively to obtain correlation analysis results, positive and negative analysis results and sentiment analysis results;
[0008] performing clustering analysis according to the correlation analysis results, the positive and negative analysis results and the sentiment analysis results to obtain an induction result corresponding to the comment information.
[0009] According to another aspect of the present disclosure, a comment information processing device is provided, comprising:
[0010] an obtaining module configured to obtain comment information, the comment information comprising a comment object, a comment content and video content corresponding to the comment content;
[0011] a first analysis module configured to perform correlation analysis, positive and negative analysis and sentiment analysis on the comment information respectively to obtain correlation analysis results, positive and negative analysis results and sentiment analysis results;
[0012] The second analysis module is configured to perform clustering analysis according to the correlation analysis result, the positive and negative analysis result, and the sentiment analysis result, and obtain an induction result corresponding to the comment information.
[0013] According to a third aspect of the present disclosure, an electronic device is provided, comprising:
[0014] at least one processor; and
[0015] a memory in communication with the at least one processor; wherein
[0016] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of the above technical solutions.
[0017] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to perform the method of any one of the above technical solutions.
[0018] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the method of any one of the above technical solutions.
[0019] The present disclosure provides a comment information processing method, device, equipment and storage medium. The present disclosure performs correlation analysis, positive and negative analysis and sentiment analysis on comment information respectively, analyzes the comment information from multiple dimensions to obtain a more comprehensive analysis result, and finally performs clustering analysis according to the obtained analysis result, so as to obtain a more comprehensive and accurate induction result corresponding to the comment information, and then the user can quickly master the key information in the comment information through the induction result.
[0020] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present disclosure, nor to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS
[0021] The accompanying drawings serve to better understand the present scheme and do not limit the present disclosure. Among them:
[0022] Figure 1 is a step schematic diagram of the comment information processing method in the embodiments of the present disclosure;
[0023] Figure 2 is a flowchart corresponding to the comment information processing method in an embodiment of the present disclosure;
[0024] Figure 3 is a flowchart corresponding to a method for processing comment information in another embodiment of the present disclosure;
[0025] Figure 4 A functional block diagram of a comment information processing device according to an embodiment of the present disclosure;
[0026] Figure 5 It is a block diagram of an electronic device used to implement the comment information processing method according to an embodiment of the present disclosure. DETAILED DESCRIPTION
[0027] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.
[0028] This disclosure provides a method for processing comment information. Figure 1 Shown, including:
[0029] Step S101: Obtain comment information, which includes a comment object, comment content, and video content corresponding to the comment content.
[0030] Specifically, the comment information is, for example, a comment related to a short video, wherein the comment information includes at least the comment object, the comment content, and the video content corresponding to the comment content. Of course, the comment information may also include other information, which is not limited here. Among them, the comment object refers to the user who posted the comment, the comment content refers to the comment text, image, expression and other content posted by the comment object, and the video content corresponding to the comment content refers to the specific video clip in which the comment object posted the comment content. To facilitate understanding of the content of the comment information, an example is given below. For example, for video A, user B posted a comment message: "Interesting". It can be obtained that the comment object in the comment information is: user B, the comment content is: "Interesting", and the video content corresponding to the comment content is: video A. For video A, other users also posted comments, and the corresponding comment information was obtained in a similar way.
[0031] Step S102 , performing relevance analysis, positive and negative analysis, and sentiment analysis on the comment information to obtain relevance analysis results, positive and negative analysis results, and sentiment analysis results.
[0032] Specifically, the relevance analysis of the comment information refers to analyzing the relevance between the comment content in the comment information and the video content corresponding to the comment content, to analyze the relevance between the two, that is, to obtain the relevance between the comment content and the video content. For example, for a certain video segment, although there are user interactions and comments below, the user's published comment content is irrelevant to the video content, and at this time the relevance analysis result can be obtained as low relevance. If the user's published comment content matches the tag content of the video segment, it can be considered that the relevance between the comment content and the video content is high.
[0033] The positive and negative analysis of the comment information refers to analyzing whether the comment content is a positive comment or a negative comment. The so-called positive comment is a positive comment content, which can include approval, agreement, support, etc. The negative comment is a negative comment content, which can include opposition, denial, etc.
[0034] In addition, the sentiment analysis of the comment information is to analyze the emotional tendency of the comment content, such as analyzing whether the comment content expresses joy or anger, surprise or melancholy, etc. By performing sentiment analysis on the comment information, the emotional tendency expressed in the comment information can be mastered.
[0035] In this way, by performing relevance analysis, positive and negative analysis, and sentiment analysis on the comment information respectively, the corresponding relevance analysis result, positive and negative analysis result, and sentiment analysis result can be obtained, which is conducive to obtaining a more comprehensive analysis result.
[0036] Step S103: performing clustering analysis according to the relevance analysis result, the positive and negative analysis result, and the sentiment analysis result, to obtain an induction result corresponding to the comment information.
[0037] Specifically, cluster analysis is an iterative cluster analysis algorithm. The general steps are: divide the obtained data into multiple groups, randomly select an object in each group as the cluster center of the group, then calculate the distance between the cluster center and other objects in each group, obtain a new grouping, and randomly select a new cluster center based on the new grouping, and so on, until the cluster center does not change, or the error square sum is locally minimized, and save the current cluster center and the group to which the cluster center belongs. In the embodiment of the present disclosure, the correlation analysis results, the positive and negative analysis results, and the sentiment analysis results are divided into different groups, and then clustered, so that each comment information can be classified, so that the final summary result can be obtained. In the embodiment of the present disclosure, when performing cluster analysis based on the correlation analysis results, the positive and negative analysis results, and the sentiment analysis results, the k-means clustering algorithm (k-means clustering algorithm, K-means clustering algorithm), DBSCAN clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, density-based clustering algorithm), or the expectation maximization clustering algorithm using a Gaussian mixture model, etc. In this way, cluster analysis and induction can help obtain more comprehensive and accurate summary results corresponding to the comment information.
[0038] The present disclosure provides a comment information processing method, apparatus, device and storage medium. The present disclosure analyzes the comment information through multiple dimensions by performing correlation analysis, positive and negative analysis and sentiment analysis on the comment information to obtain a more comprehensive analysis result. Finally, a cluster analysis is performed based on the obtained analysis results, which is conducive to obtaining a more comprehensive and accurate summary result corresponding to the comment information, and further enables users to quickly grasp the key information in the comment information through the summary result.
[0039] In some optional embodiments, the comment information is subjected to relevance analysis, positive and negative analysis, and sentiment analysis, respectively, to obtain relevance analysis results, positive and negative analysis results, and sentiment analysis results, including:
[0040] Performing correlation analysis on the comment content in the comment information and the video content corresponding to the comment content to obtain a correlation analysis result;
[0041] Performing positive and negative analysis on the comments made by the comment subjects in the comment information regarding the video content to obtain positive and negative analysis results;
[0042] The sentiment analysis is performed on the comments made by the comment subjects in the comment information regarding the video content to obtain the sentiment analysis results.
[0043] Specifically, when the correlation analysis is performed on the comment information, the comment content in the comment information and the video content corresponding to the comment content are analyzed for correlation, and a correlation analysis result is obtained. The correlation analysis result shows the correlation between the comment content and the video content. When the comment content in the comment information and the video content corresponding to the comment content are analyzed for correlation, the text information of the comment content and the keywords corresponding to the video content can be extracted, and then the similarity between the comment content and the video content is calculated. According to the similarity between the two, the correlation between the comment content and the video content can be obtained, and then the correlation analysis result can be obtained.
[0044] When the positive and negative analysis is performed on the comment information, the comment content made by the comment object to the video content in the comment information is analyzed for positive and negative, and a positive and negative analysis result is obtained. When the comment content made by the comment object to the video content in the comment information is analyzed for positive and negative, the comment content can be preprocessed (such as word segmentation, keyword extraction, etc.) first, and then the preprocessed data is classified according to a binary classification algorithm, so that the positive and negative analysis result can be obtained. The binary classification algorithm can be, for example, logistic regression, support vector machine, and naive Bayes algorithm.
[0045] When the sentiment analysis is performed on the comment information, the comment content made by the comment object to the video content in the comment information is analyzed for sentiment, and a sentiment analysis result is obtained. When the comment content made by the comment object to the video content in the comment information is analyzed for sentiment, a deep learning-based sentiment analysis method can be used. The deep learning-based sentiment analysis method is performed using a neural network. Typical neural network learning methods include convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory (LSTM) network, etc. When performing sentiment analysis, a dictionary-based sentiment analysis method can also be used to obtain a sentiment analysis result. The dictionary-based sentiment analysis method specifically constructs a sentiment dictionary, which is a dictionary containing a large number of words and their sentiment polarity (positive, negative, or neutral). The words in the text corresponding to the comment content are matched with the sentiment, so as to determine the sentiment tendency of the text. When performing sentiment analysis, only the words in the text are matched with the sentiment dictionary, so as to determine the sentiment tendency of the text, and then the sentiment analysis result can be obtained.
[0046] In this way, the comment information is analyzed in multiple dimensions through correlation analysis, positive and negative aspect analysis, and sentiment analysis, so that a more comprehensive analysis result is obtained.
[0047] In some optional embodiments, clustering analysis is performed according to the correlation analysis result, the positive and negative aspect analysis result, and the sentiment analysis result, to obtain an induction result corresponding to the comment information, including:
[0048] The correlation analysis result, the positive and negative aspect analysis result, and the sentiment analysis result are respectively subjected to clustering analysis to obtain a clustering result corresponding to the correlation analysis result, a clustering result corresponding to the positive and negative aspect analysis result, and a clustering result corresponding to the sentiment analysis result.
[0049] According to the clustering result corresponding to the correlation analysis result, the clustering result corresponding to the positive and negative aspect analysis result, and the clustering result corresponding to the sentiment analysis result, an induction result corresponding to the comment information is obtained.
[0050] Specifically, before obtaining the induction result corresponding to the comment information, the correlation analysis result, the positive and negative aspect analysis result, and the sentiment analysis result are respectively subjected to clustering analysis to obtain a clustering result corresponding to the correlation analysis result, a clustering result corresponding to the positive and negative aspect analysis result, and a clustering result corresponding to the sentiment analysis result, and then the clustering results are integrated to obtain the induction result corresponding to the comment information. In the clustering analysis of each analysis result, a k-means clustering algorithm, a DBSCAN clustering algorithm, or an expectation maximization clustering algorithm of a Gaussian mixture model can be used. Since the implementation of the clustering algorithm of the present disclosure is similar to that of the prior art, it will not be described here. In addition, the same clustering algorithm or different clustering algorithms can be used in the clustering analysis of each analysis result.
[0051] In this way, each analysis result is clustered first, and then the clustering results are integrated, so that the induction result obtained is more comprehensive. In addition, by separately analyzing each analysis result and then integrating the analysis results, the amount of data processed each time is moderate, which is conducive to improving the data processing efficiency.
[0052] In some optional embodiments, the clustering analysis is respectively performed on the correlation analysis result, the positive and negative aspect analysis result, and the sentiment analysis result, to obtain a clustering result corresponding to the correlation analysis result, a clustering result corresponding to the positive and negative aspect analysis result, and a clustering result corresponding to the sentiment analysis result, including:
[0053] The correlation analysis result is classified and summarized to obtain the clustering result corresponding to the correlation analysis result.
[0054] The positive and negative aspect analysis result is classified and summarized to obtain the clustering result corresponding to the positive and negative aspect analysis result.
[0055] The sentiment analysis result is classified and summarized to obtain the clustering result corresponding to the sentiment analysis result.
[0056] Specifically, when classifying and summarizing the correlation analysis result, the k-means algorithm can be used for classification and summarization. Through the clustering analysis algorithm, the comment information with high correlation can be clustered into a class, and each sample is given a label using a label tree, so that the clustering result corresponding to the correlation analysis result can be obtained.
[0057] When classifying and summarizing the positive and negative aspect analysis result, the DBSCAN clustering algorithm can be used to perform density clustering processing on the positive analysis result and the negative analysis result in the positive and negative aspect analysis result, so that the clustering result corresponding to the positive and negative aspect analysis result can be obtained.
[0058] When classifying and summarizing the sentiment analysis result, the k-means algorithm can be used. Through the clustering analysis algorithm, samples with similar sentiment characteristics can be clustered into a class, and each sample is given a label using a label tree, so that the clustering result corresponding to the sentiment analysis result can be obtained.
[0059] In this way, by respectively performing clustering analysis on the correlation analysis result, the positive and negative aspect analysis result, and the sentiment analysis result, and by classifying each analysis result, it is beneficial to master more comprehensive information in the comment information.
[0060] In some optional embodiments, after obtaining the induction result corresponding to the comment information, the method further includes:
[0061] According to a preset filtering strategy, the comment information satisfying the strategy is filtered from the induction result corresponding to the comment information and is displayed.
[0062] Specifically, after obtaining the summary result corresponding to the comment information, a manually defined screening strategy can be adopted, and the screening strategy clearly defines the screening conditions. After determining the screening strategy, it is convenient to screen the comment information meeting the strategy from the summary result corresponding to the comment information. After obtaining the comment information meeting the strategy, the comment information is displayed, so that the comments of interest to the user can be displayed.
[0063] In this way, after obtaining the summary result corresponding to the comment information, the comment information meeting the strategy is screened from the summary result corresponding to the comment information according to the preset screening strategy and displayed, thereby facilitating the user to quickly locate the comment information to be browsed.
[0064] In some optional embodiments, the screening strategy includes at least one of the following:
[0065] According to the historical interaction data of the user, the user interest point is determined, and the comment information associated with the user interest point is extracted;
[0066] The comment information interacted by the user is extracted;
[0067] The comment information published by the user in real time and passed by the audit is extracted;
[0068] The comment information meeting the user-defined is extracted.
[0069] Specifically, in the related technical solution, when the number of comments is large, the user has difficulty in quickly finding the comments of interest, that is, there is a lack of effective screening and search function, making it difficult for the user to find specific information in the comments. In order to solve the above problem, the present disclosure determines the user interest point according to the historical interaction data of the user, and extracts the comment information associated with the user interest point, that is, the present disclosure extracts the comment content based on the user's interest, combines the user's comment browsing habits and interaction operations to generate interest points, and then extracts the comment content according to the relevance of the interest points, for the user to preview and consume, thereby facilitating to improve the user experience.
[0070] On the other hand, in the related technical solution, most of the existing comment systems are static, lacking real-time social interaction, making the replies and discussions between users not intuitive enough, and it is difficult to form an interesting discussion atmosphere. In order to solve the above problem, the present application extracts the comment information interacted by the user, extracts the comment content participated by the user in various ways such as likes and replies from the user latitude, for the user to preview or continue to operate, thereby enriching the form of interaction.
[0071] In addition, in the related technical solutions, since some platforms allow anonymous comments, this can lead to the emergence of some inappropriate remarks, reducing the quality of the comment area and increasing the possibility of the spread of false information. To solve the above problems, the present application extracts real-time comment information published by users and passed by the audit, real-time comment content extraction, for presenting real-time comment content extraction published by other users in real time and passed by the audit, for users to obtain real-time comment content.
[0072] Finally, the present disclosure also supports customized comment content filtering, supports extracting comment content according to user descriptions, and supports generating specific comment content according to related descriptions and one-key publishing.
[0073] In this way, by setting different filtering strategies, the interactive form can be enriched, the interest and interactivity of the video can be improved, the consumption of the user comment interactive scene can be improved, the creative thinking of the user can be stimulated, and the interactive form of the video can be enriched.
[0074] In order to facilitate the overall understanding of the technical solutions of the present disclosure, see Figure 2 , Figure 2 is a flowchart of a comment information processing method in an embodiment of the present disclosure. The method comprises:
[0075] S201: Obtain comment information, the comment information comprising a comment publisher, comment content, and associated video content.
[0076] Specifically, the comment publisher can also be referred to as a comment object, the comment publisher refers to a user who publishes a comment, the comment content refers to comment text, images, expressions, and other content published by the comment object, and the video content corresponding to the comment content refers to a specific video segment in which the comment object publishes the comment content.
[0077] S202: Perform relevance analysis, positive and negative analysis, and sentiment analysis on the comment content respectively to obtain relevance analysis results, positive and negative analysis results, and sentiment analysis results.
[0078] Specifically, in the relevance analysis of the comment content, the text information of the comment content and the keywords corresponding to the video content can be extracted, and then the similarity between the comment content and the video content is calculated, and the relevance between the comment content and the video content can be obtained according to the similarity between them, and then the relevance analysis result can be obtained. In the positive and negative analysis of the comment content, the comment content can be preprocessed (such as word segmentation, keyword extraction, etc.) first, and then the preprocessed data is classified according to the binary classification algorithm, so that the positive and negative analysis result can be obtained. The binary classification algorithm can be, for example, logistic regression, support vector machine, and naive Bayes algorithm. In the sentiment analysis of the comment content, a deep learning sentiment analysis method can be used. The deep learning sentiment analysis method is performed using a neural network. Typical neural network learning methods include convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory (LSTM) network, etc.
[0079] S203: According to the relevance analysis result, the positive and negative analysis result, and the sentiment analysis result, the weight score clustering output is integrated.
[0080] Specifically, after obtaining the relevance analysis result, the positive and negative analysis result, and the sentiment analysis result, the relevance analysis result, the positive and negative analysis result, and the sentiment analysis result are respectively clustered and analyzed to obtain the clustering result corresponding to the relevance analysis result, the clustering result corresponding to the positive and negative analysis result, and the clustering result corresponding to the sentiment analysis result. In the clustering analysis of each analysis result, a k-means clustering algorithm, a DBSCAN clustering algorithm (Density-Based Spatial Clustering of Applications with Noise), or an expectation maximization clustering algorithm using a Gaussian mixture model can be used.
[0081] S204: Give an induction summary according to the clustering of each comment.
[0082] Finally, the various clustering results are integrated to obtain the induction results corresponding to the comment information. In this way, the various analysis results are clustered first, and then the various aggregation results are integrated, so that the obtained induction results are more comprehensive. In addition, by separately analyzing the various analysis results and then integrating the analysis results, the amount of data processed each time is moderate, which is conducive to improving the data processing efficiency.
[0083] In addition, referring to Figure 3 , Figure 3 is a flowchart of a comment information processing method in another embodiment of the present disclosure. The method flow includes:
[0084] S301: Obtain all comments.
[0085] S302: Input all comments into an intelligent analysis model for analysis to obtain a comment induction summary.
[0086] S303: Perform intelligent reading mode based on the comment induction summary, comment content extraction based on user interest, comment content extraction based on user participation, real-time comment content extraction, and comment content filtering based on customization.
[0087] Since the flow of the method is similar to the content of the above-mentioned embodiments, for detailed explanation thereof, reference can be made to the content of the foregoing embodiments, which will not be described herein.
[0088] The device embodiments of the present application are described below, which can be used to execute the comment information processing method in the above-mentioned embodiments of the present application. For details not disclosed in the device embodiments of the present application, reference can be made to the above-mentioned embodiments of the comment information processing method.
[0089] The present disclosure also provides a comment information processing device 400, as shown in Figure 4 , comprising:
[0090] The acquisition module 401 is configured to acquire comment information, the comment information including a comment object, comment content, and video content corresponding to the comment content.
[0091] The first analysis module 402 is configured to perform correlation analysis, positive and negative aspect analysis, and sentiment analysis on the comment information respectively to obtain correlation analysis results, positive and negative aspect analysis results, and sentiment analysis results.
[0092] The second analysis module 403 is configured to perform clustering analysis according to the correlation analysis results, the positive and negative aspect analysis results, and the sentiment analysis results to obtain induction results corresponding to the comment information.
[0093] In some optional embodiments, the first analysis module 402 performs relevance analysis, positive and negative analysis, and sentiment analysis on the comment information, and obtains relevance analysis results, positive and negative analysis results, and sentiment analysis results, including:
[0094] Performing correlation analysis on the comment content in the comment information and the video content corresponding to the comment content to obtain a correlation analysis result;
[0095] Performing positive and negative analysis on the comments made by the comment subjects in the comment information regarding the video content to obtain positive and negative analysis results;
[0096] The sentiment analysis is performed on the comments made by the comment subjects in the comment information regarding the video content to obtain the sentiment analysis results.
[0097] In some optional embodiments, the second analysis module 403 performs cluster analysis based on the correlation analysis results, the positive and negative analysis results, and the sentiment analysis results to obtain summary results corresponding to the comment information, including:
[0098] Perform cluster analysis on the correlation analysis results, the positive and negative analysis results, and the sentiment analysis results respectively, to obtain cluster results corresponding to the correlation analysis results, cluster results corresponding to the positive and negative analysis results, and cluster results corresponding to the sentiment analysis results;
[0099] According to the clustering results corresponding to the correlation analysis results, the clustering results corresponding to the positive and negative analysis results, and the clustering results corresponding to the sentiment analysis results, the summary results corresponding to the comment information are obtained.
[0100] In some optional embodiments, the second analysis module 403 performs cluster analysis on the correlation analysis results, the positive and negative analysis results, and the sentiment analysis results, respectively, to obtain cluster results corresponding to the correlation analysis results, cluster results corresponding to the positive and negative analysis results, and cluster results corresponding to the sentiment analysis results, including:
[0101] Classify and summarize the correlation analysis results to obtain clustering results corresponding to the correlation analysis results;
[0102] Classify and summarize the positive and negative analysis results to obtain clustering results corresponding to the positive and negative analysis results;
[0103] The sentiment analysis results are classified and summarized to obtain clustering results corresponding to the sentiment analysis results.
[0104] In some optional embodiments, after obtaining the summary results corresponding to the comment information, the second analysis module 403 is further configured to filter the comment information that meets the strategy from the summary results corresponding to the comment information according to a preset filtering strategy and display the filtered comment information.
[0105] In some optional embodiments, the screening strategy comprises at least one of the following:
[0106] determining a user interest point according to historical interaction data of the user, and extracting comment information associated with the user interest point;
[0107] extracting comment information that the user interacts with;
[0108] extracting comment information published by the user in real time and approved;
[0109] extracting comment information that meets a user-defined criterion.
[0110] In the technical solutions of the present disclosure, the acquisition, storage and application of user personal information comply with relevant laws and regulations and do not violate public order and good customs.
[0111] According to the embodiments of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium and a computer program product.
[0112] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit implementations of the present disclosure described and / or claimed in this document.
[0113] As shown in Figure 5 The electronic device 500 includes a computing unit 501 that can perform various appropriate actions and processes according to computer programs stored in a read-only memory (ROM) 502 or loaded into a random access memory (RAM) 503 from a storage unit 508. Various programs and data required for the operation of the device 500 can also be stored in the RAM 503. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.
[0114] A plurality of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a magnetic disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.
[0115] The computing unit 501 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 501 performs various methods and processes described above, such as the review information processing method. For example, in some embodiments, the review information processing method can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as the storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded onto the RAM 503 and executed by the computing unit 501, one or more steps of the above-described microprogram distribution can be performed. Alternatively, in other embodiments, the computing unit 501 can be configured to perform the review information processing method by any other appropriate means, such as by means of firmware.
[0116] Various implementations of the systems and techniques described above herein can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.
[0117] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a standalone software package, or entirely on a remote machine or server.
[0118] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0119] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.
[0120] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.
[0121] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.
[0122] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.
[0123] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.
Claims
1. A method for processing comment information, comprising: Obtaining comment information, wherein the comment information includes the comment object, comment content, and video content corresponding to the comment content; Performing correlation analysis, positive and negative analysis, and sentiment analysis on the comment information to obtain correlation analysis results, positive and negative analysis results, and sentiment analysis results; A cluster analysis is performed based on the correlation analysis result, the positive and negative analysis result, and the sentiment analysis result to obtain a summary result corresponding to the comment information.
2. The method according to claim 1, wherein The relevance analysis, positive and negative analysis, and sentiment analysis are performed on the comment information to obtain the relevance analysis results, positive and negative analysis results, and sentiment analysis results, including: Performing correlation analysis on the comment content in the comment information and the video content corresponding to the comment content to obtain a correlation analysis result; Performing positive and negative analysis on the comments made by the comment subject in the comment information regarding the video content to obtain positive and negative analysis results; The sentiment analysis is performed on the comments made by the comment subjects in the comment information regarding the video content to obtain a sentiment analysis result.
3. The method according to claim 1, wherein The cluster analysis is performed based on the correlation analysis result, the positive and negative analysis result, and the sentiment analysis result to obtain a summary result corresponding to the comment information, including: Performing cluster analysis on the correlation analysis results, the positive and negative analysis results, and the sentiment analysis results respectively to obtain cluster results corresponding to the correlation analysis results, cluster results corresponding to the positive and negative analysis results, and cluster results corresponding to the sentiment analysis results; According to the clustering results corresponding to the relevance analysis results, the clustering results corresponding to the positive and negative analysis results, and the clustering results corresponding to the sentiment analysis results, a summary result corresponding to the comment information is obtained.
4. The method according to claim 3, wherein: The cluster analysis is performed on the correlation analysis result, the positive and negative analysis result, and the sentiment analysis result respectively to obtain a cluster result corresponding to the correlation analysis result, a cluster result corresponding to the positive and negative analysis result, and a cluster result corresponding to the sentiment analysis result, including: Classifying and summarizing the correlation analysis results to obtain clustering results corresponding to the correlation analysis results; Classifying and summarizing the positive and negative analysis results to obtain clustering results corresponding to the positive and negative analysis results; The sentiment analysis results are classified and summarized to obtain clustering results corresponding to the sentiment analysis results.
5. The method according to claim 1, wherein After obtaining the summary result corresponding to the comment information, the method further includes: According to a preset screening strategy, the review information that meets the strategy is screened from the summary results corresponding to the review information and displayed.
6. The method according to claim 5, wherein: The screening strategy includes at least one of the following: Determine user interest points based on the user's historical interaction data, and extract comment information associated with the user interest points; Extract comment information of user interactions; Extract the review information published by users in real time and approved; Extract user-defined comment information.
7. A comment information processing device, comprising: An acquisition module is used to acquire comment information, wherein the comment information includes a comment object, comment content, and video content corresponding to the comment content; The first analysis module is used to perform relevance analysis, positive and negative analysis, and sentiment analysis on the comment information to obtain relevance analysis results, positive and negative analysis results, and sentiment analysis results; The second analysis module is used to perform cluster analysis based on the correlation analysis result, the positive and negative analysis result, and the sentiment analysis result to obtain a summary result corresponding to the comment information.
8. The device according to claim 7, wherein The first analysis module performs relevance analysis, positive and negative analysis, and sentiment analysis on the comment information, and obtains relevance analysis results, positive and negative analysis results, and sentiment analysis results, including: Performing correlation analysis on the comment content in the comment information and the video content corresponding to the comment content to obtain a correlation analysis result; Performing positive and negative analysis on the comments made by the comment subject in the comment information regarding the video content to obtain positive and negative analysis results; The sentiment analysis is performed on the comments made by the comment subjects in the comment information regarding the video content to obtain a sentiment analysis result.
9. The device according to claim 7, wherein The second analysis module performs cluster analysis based on the correlation analysis result, the positive and negative analysis result, and the sentiment analysis result to obtain a summary result corresponding to the comment information, including: Performing cluster analysis on the correlation analysis results, the positive and negative analysis results, and the sentiment analysis results respectively to obtain cluster results corresponding to the correlation analysis results, cluster results corresponding to the positive and negative analysis results, and cluster results corresponding to the sentiment analysis results; According to the clustering results corresponding to the relevance analysis results, the clustering results corresponding to the positive and negative analysis results, and the clustering results corresponding to the sentiment analysis results, a summary result corresponding to the comment information is obtained.
10. The device according to claim 9, wherein The second analysis module performs cluster analysis on the correlation analysis result, the positive and negative analysis result, and the sentiment analysis result, respectively, to obtain cluster results corresponding to the correlation analysis result, cluster results corresponding to the positive and negative analysis result, and cluster results corresponding to the sentiment analysis result, including: Classifying and summarizing the correlation analysis results to obtain clustering results corresponding to the correlation analysis results; Classifying and summarizing the positive and negative analysis results to obtain clustering results corresponding to the positive and negative analysis results; The sentiment analysis results are classified and summarized to obtain clustering results corresponding to the sentiment analysis results.
11. The device according to claim 7, wherein After obtaining the summary results corresponding to the comment information, the second analysis module is further configured to filter out comment information that meets the strategy from the summary results corresponding to the comment information according to a preset screening strategy and display the filtered comment information.
12. The device according to claim 11, wherein The screening strategy includes at least one of the following: Determine user interest points based on the user's historical interaction data, and extract comment information associated with the user interest points; Extract comment information of user interactions; Extract the review information published by users in real time and approved; Extract user-defined comment information.
13. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.
14. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 6.
15. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.