CN-115374307-B - Intelligent public opinion analysis method, device, terminal equipment and storage medium

CN115374307BCN 115374307 BCN115374307 BCN 115374307BCN-115374307-B

Abstract

The invention discloses an intelligent public opinion analysis method, device, terminal equipment and storage medium, wherein the method comprises the steps of collecting video barrage data of a user; generating a focused video image based on the acquired user video barrage data, generating an intelligent public opinion primary analysis result based on the generated focused video image, generating barrage analysis information based on the acquired user video barrage data and the intelligent public opinion primary analysis result, and carrying out video public opinion analysis based on the intelligent public opinion primary analysis result and barrage analysis information. Compared with the traditional method, the intelligent public opinion analysis method based on barrage acquisition is lower in cost, more complete in coverage, higher in efficiency and higher in accuracy.

Inventors

HE RUIXIN

Assignees

中国移动通信集团设计院有限公司
中国移动通信集团有限公司

Dates

Publication Date: 20260508
Application Date: 20210518

Claims (6)

1. An intelligent public opinion analysis method is characterized by comprising the following steps: collecting video bullet screen data of a user; generating a concerned video portrait based on the collected user video barrage data, and generating an intelligent public opinion primary analysis result based on the generated concerned video portrait; the method comprises the steps of training classified video description models based on the collected user video bullet screen data to obtain video description models aiming at different classes, establishing a video portrait model based on the video description models, generating a concerned video portrait based on the video portrait model, and generating an intelligent public opinion preliminary analysis result based on the generated concerned video portrait, wherein the video portrait model comprises portrait dimension, video portrait labels and label weights; The method comprises the steps of collecting user video barrage data and intelligent public opinion preliminary analysis results, generating barrage analysis information based on the collected user video barrage data and the intelligent public opinion preliminary analysis results, reading corresponding user video barrage data for videos in the public opinion monitoring video list, generating barrage basic attribute analysis results based on the read user video barrage data, generating barrage emotion analysis information, selecting barrages with emotion value scores lower than a preset threshold value based on the barrage emotion analysis information, generating negative barrage user portrait results, integrating the generated barrage basic attribute analysis results, barrage emotion analysis information and negative barrage user portrait results, and generating barrage analysis information; Performing video public opinion analysis based on the intelligent public opinion preliminary analysis result and bullet screen analysis information; Wherein generating a video representation of interest based on the video representation model, the step of generating an intelligent public opinion preliminary analysis result based on the generated video representation of interest comprises: obtaining a public opinion monitoring coefficient of the concerned video based on the label weight of the video portrait model; Selecting a preset number of videos with the public opinion monitoring coefficients ranked at the front as the focused video images and classifying the focused video images to generate corresponding video description information; Generating a public opinion monitoring video list based on the generated corresponding video description information, and taking the public opinion monitoring video list as an intelligent public opinion primary analysis result; Based on the barrage emotion analysis information, selecting a barrage with emotion value scores lower than a preset threshold value, and generating a negative barrage user portrait result comprises the following steps: based on the barrage emotion analysis information, reading video barrage data of the negative barrage users with emotion value scores lower than a preset threshold value; Correlating the read video barrage data of the negative barrage user with a pre-established user scene information base to obtain a barrage user positioning database, wherein the user scene information base is established based on automatically collecting the video barrage data of the user and correlating XDR data, MDT data, MR data, industrial parameter data and geographical scene frame data; And carrying out fine portrayal on the negative barrage users focused by the analysis and positioning based on the barrage user positioning database, and generating negative barrage user portrayal results.
2. The intelligent public opinion analysis method of claim 1, wherein the step of generating a public opinion monitoring video list based on the generated corresponding video description information further comprises: comparing corresponding video description information for each concerned video image corresponding to the public opinion monitoring video list to obtain the similarity association degree of each video; And reserving the video with the highest public opinion monitoring coefficient for a plurality of videos with the similarity relevance higher than a set threshold value, deleting other similar videos, sequentially selecting complementary videos from the videos with the preset quantity according to the public opinion monitoring coefficient until the public opinion monitoring video list is not changed, and obtaining a final public opinion monitoring video list as an intelligent public opinion preliminary analysis result.
3. The intelligent public opinion analysis method according to any one of claims 1-2, wherein the step of performing video public opinion analysis based on the intelligent public opinion preliminary analysis result and barrage analysis information comprises: calculating to obtain important public opinion processing priority based on the public opinion monitoring video list and the public opinion monitoring coefficients of the corresponding videos; Forming a comprehensive intelligent public opinion analysis scheme based on the intelligent public opinion primary analysis result and the barrage analysis information and outputting and/or pushing the public opinion analysis result to the outside; and/or processing videos with priorities reaching a preset threshold value, and realizing the geographical presentation of public opinion scenes based on a pre-established user scene information base.
4. An intelligent public opinion analysis device, its characterized in that, intelligent public opinion analysis device includes: The acquisition module is used for acquiring video barrage data of a user; The list generation module is used for generating a concerned video portrait based on the collected user video bullet screen data and generating an intelligent public opinion primary analysis result based on the generated concerned video portrait; the list generation module is also used for training classified video description models based on the collected user video barrage data to obtain video description models for different types; generating a video portrait model based on the video description model, generating an intelligent public opinion preliminary analysis result based on the generated video portrait, wherein the video portrait model comprises portrait dimension, video portrait label and label weight; The list generation module is also used for obtaining public opinion monitoring coefficients of the concerned videos based on the tag weight of the video portrait model, selecting a preset number of videos with the top ranking of the public opinion monitoring coefficients as the concerned video portraits and classifying to generate corresponding video description information; The information generation module is used for generating barrage analysis information based on the acquired user video barrage data and the intelligent public opinion preliminary analysis result, reading corresponding user video barrage data for videos in the public opinion monitoring video list, generating barrage basic attribute analysis results based on the read user video barrage data, generating barrage emotion analysis information, selecting barrages with emotion value scores lower than a preset threshold value based on the barrage emotion analysis information, generating negative barrage user portrait results, integrating the generated barrage basic attribute analysis results, barrage emotion analysis information and negative barrage user portrait results, and generating barrage analysis information; The information generation module is also used for reading negative barrage user video barrage data with emotion value scores lower than a preset threshold value based on the barrage emotion analysis information, correlating the read negative barrage user video barrage data with a pre-established user scene information base to obtain a barrage user positioning database, carrying out fine portraits on the user concerned with the negative barrage analyzed and positioned based on the barrage user positioning database to generate a negative barrage user portrait result, wherein the user scene information base is established based on automatically collecting user video barrage data, correlating XDR data, MDT data, MR data, industrial parameter data and geographic scene frame data; and the analysis module is used for carrying out video public opinion analysis based on the intelligent public opinion preliminary analysis result and the barrage analysis information.
5. A terminal device, characterized in that it comprises a memory, a processor and a computer program stored on the memory and executable on the processor, which computer program, when executed by the processor, implements the intelligent public opinion analysis method according to any of claims 1-3.
6. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, which when executed by a processor, implements the intelligent public opinion analysis method of any of claims 1-3.

Description

Intelligent public opinion analysis method, device, terminal equipment and storage medium Technical Field The present invention relates to the field of mobile communications technologies, and in particular, to an intelligent public opinion analysis method, device, terminal equipment, and storage medium. Background At present, mining analysis processing of social public opinion information mainly relies on user investigation and traditional text analysis means, and comment conditions are browsed through platforms such as questionnaires, microblogs, bar sticking and knowledge, comment text keywords are identified and counted, and public opinion analysis functions such as event tag classification and hotword search recommendation are realized. The existing analytical schemes suffer from the following disadvantages: For the traditional text analysis method, the comments of platform users such as microblogs, forums and the like are not obvious in time characteristics and lack of real-time interactivity and sharing performance compared with video barrages, and barrage texts have the characteristics of short length, more network terms, incomplete information, strong correlation with the current scene of the video and the like compared with the traditional comment texts. In addition, the traditional means are single for user characteristic images (generally only gender, approximate age, provincial area distribution and the like) and analysis, and are difficult to accurately guide public opinion analysis. Disclosure of Invention The invention mainly aims to provide an intelligent public opinion analysis method, an intelligent public opinion analysis device, terminal equipment and a storage medium, and aims to realize accurate analysis and intelligent efficient processing of social public opinion information and improve the accuracy of the public opinion information. In order to achieve the above object, an embodiment of the present invention provides an intelligent public opinion analysis method, which includes the following steps: collecting video bullet screen data of a user; generating a concerned video portrait based on the collected user video barrage data, and generating an intelligent public opinion primary analysis result based on the generated concerned video portrait; Generating barrage analysis information based on the collected user video barrage data and the intelligent public opinion preliminary analysis result; And carrying out video public opinion analysis based on the intelligent public opinion preliminary analysis result and the barrage analysis information. Optionally, the step of generating the focused video portrait based on the collected user video barrage data and generating the intelligent public opinion preliminary analysis result based on the generated focused video portrait includes: Based on the collected user video barrage data, training classified video description models to obtain video description models aiming at different types; establishing a video image model based on the video description model; generating a video image of interest based on the video image model, and generating an intelligent public opinion primary analysis result based on the generated video image of interest. Optionally, the step of generating the interesting video image based on the video image model and generating the intelligent public opinion preliminary analysis result based on the generated interesting video image comprises: Based on the video portrait model, obtaining a public opinion monitoring coefficient of the concerned video; Selecting a preset number of videos with the public opinion monitoring coefficients ranked at the front as the focused video images and classifying the focused video images to generate corresponding video description information; And generating a public opinion monitoring video list based on the generated corresponding video description information to serve as an intelligent public opinion preliminary analysis result. Optionally, the step of generating the public opinion monitoring video list based on the generated corresponding video description information further includes: comparing corresponding video description information for each concerned video image corresponding to the public opinion monitoring video list to obtain the similarity association degree of each video; And reserving the video with the highest public opinion monitoring coefficient for a plurality of videos with the similarity relevance higher than a set threshold value, deleting other similar videos, sequentially selecting complementary videos from the videos with the preset quantity according to the public opinion monitoring coefficient until the public opinion monitoring video list is not changed, and obtaining a final public opinion monitoring video list as an intelligent public opinion preliminary analysis result. Optionally, the step of generating barrage analysis information based on the collected use