Video Popularity Classification via Short Message Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video delivery services face challenges in accurately determining video popularity due to ambiguous short messages from users, which can lead to inaccurate classification and rating of video content.
Innovation Solution
A method using classifiers to select relevant accounts and tokens, normalize short messages with a mixed model combining a lexicon and translation model, and generate ratings for video content based on sentiment analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If short messages are used to determine video popularity, then user interaction and service offerings are improved, but the accuracy of content classification deteriorates due to ambiguous messaging
Solution Approach 1:
The classification task is divided into multiple stages: first identifying relevant accounts, then classifying messages at the account level, and finally rating individual messages. This segmentation allows each stage to focus on specific aspects, improving overall accuracy while maintaining user interaction benefits
Solution Approach 2:
The system performs preliminary classification to identify accounts related to a video before classifying individual messages. This preliminary action filters out irrelevant accounts and messages early in the process, improving the accuracy of subsequent classification while preserving the ability to capture user interaction signals
2Productivity
If short messages are processed without normalization, then processing speed is maintained, but classification accuracy deteriorates due to varied messaging formats
Solution Approach 1:
The system changes the parameters of short messages through normalization, transforming them into a standardized format that improves classification accuracy. The normalization process adjusts messaging parameters (formatting, capitalization, punctuation) while preserving the core meaning, enabling accurate classification without sacrificing processing efficiency
Data Source
AI summary
In one embodiment, a method selects a set of accounts related to a video using a first classifier that is trained to identify the set of accounts from a plurality of accounts and selects a token used to identify the video. A second classifier is trained to identify short messages that are related to the video using a first set of short messages from the accounts related to the video and the second set of short messages including the tokens. A third set of short messages is identified that is related to the video from a plurality of short messages. The method normalizes words in the set of short messages using a mixed model including a lexicon model and a translation model. Then, the method generates a set of ratings for the set of short messages and uses the set of ratings in a service for a video delivery service.