Unlock instant, AI-driven research and patent intelligence for your innovation.

Video Bullet Screen Sentiment Classification Method Based on Improved Bayesian Model

A Bayesian model and emotion classification technology, which is applied in character and pattern recognition, instruments, calculations, etc., can solve problems affecting algorithm judgment and achieve good classification results

Inactive Publication Date: 2021-09-28
CHENGDU UNIV OF INFORMATION TECH
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Influenced by Internet culture, the video barrage is full of Internet terminology, word meaning deformation and multiple meanings, which affect the judgment of the algorithm

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Video Bullet Screen Sentiment Classification Method Based on Improved Bayesian Model
  • Video Bullet Screen Sentiment Classification Method Based on Improved Bayesian Model
  • Video Bullet Screen Sentiment Classification Method Based on Improved Bayesian Model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0048] 1. In order to facilitate further understanding, the following first analyzes the characteristics of the video barrage:

[0049] 1.1. Real-time performance of video barrage:

[0050] Video barrage is a new form of user comments, which is different from traditional user comments because of its high real-time nature. The real-time performance of the video barrage is as follows: the evaluation object of the video barrage is the content of the video at that time (such as the characters, events or plots in the video at a certain moment). As the playing time goes by, the video content changes constantly. Changes have led to changes in the evaluation objects and emotional tendencies of video barrage. Therefore, even if they come from the same video, video bullet chats with different publication times have different evaluation objects, and their emotional tendencies may be completely opposite.

[0051] In terms of sentiment analysis, sentiment information usually consists of ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a video barrage emotion classification method based on an improved Bayesian model, comprising the following steps: using a clustering algorithm to divide the video barrage into time periods, and determining parameters q and v, where q is the replacement rate, and v is Target correction rate; use the Bayesian model to initially classify the data set, record the period category probability θ and prior probability ω; use ω+q(θ‑ω) to replace the prior probability ω, reclassify the sample, calculate and record The correction amount d and its variation Δd are used to update the period category probability θ; if Δd≠0, return to the third step; select the correction result with the correction rate closest to v as the final result. Combining the real-time characteristics of video barrage, the present invention proposes a corrected Bayesian algorithm based on period division, that is, an improved Bayesian model, and uses it in the emotional classification of video barrage, correcting the traditional Bayesian model error. It has a good classification effect and is especially suitable for video barrage with high consistency of emotional tendency.

Description

technical field [0001] The invention relates to a video bullet chat emotion classification method, in particular to a video bullet chat emotion classification method based on an improved Bayesian model. Background technique [0002] Now that we have entered the era of Web 2.0, the rapid development of Internet technology and social media has given birth to a new form of comment - video barrage. Video barrage refers to a technology that appears on the video player to enable user comments to be played synchronously with the video. As soon as the video barrage appeared, it aroused the welcome of netizens, and its usage remained high. Influenced by its Tucao culture, video barrage contains a lot of emotional information. At the same time, its high real-time appearance is also different from traditional user comments. Therefore, sentiment analysis on video barrage has high research value. [0003] The main content of sentiment analysis can be summarized as the extraction of em...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62
CPCG06F18/23213G06F18/24155G06F18/214
Inventor 安俊秀王梓懿靳宇倡刘敦虎
Owner CHENGDU UNIV OF INFORMATION TECH