Message Recommendation via Topic Model Intermediary
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current social media platforms lack flexibility in message delivery, often pushing irrelevant messages to users while failing to provide interested messages from unfollowed users.
Innovation Solution
A method and apparatus that parse and compare interest and topic description information between users' messages using probability distribution analysis, employing Gibbs sampling and similarity algorithms to recommend messages based on user interest, ensuring relevant content is pushed to users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If messages are pushed based on user's followed circle, then users can obtain messages from followed users, but users cannot obtain interesting messages from unfollowed users and receive many irrelevant messages
Solution Approach 1:
The patent introduces a topic model as an intermediary between users and messages. Instead of directly relying on social relationships (followed circle), the system uses topic distribution as a mediator to match users with messages. The topic model extracts thematic information from messages and compares it with user interests, enabling users to receive relevant messages from unfollowed users while filtering out irrelevant content.
2Reliability
If messages are pushed based on user's followed circle, then users can obtain messages from followed users, but users receive many irrelevant messages wasting time and energy
Solution Approach 1:
The patent replaces the mechanical social relationship system (follow/unfollow) with a semantic matching system based on topic models. Instead of relying on social graph connections, the system uses probabilistic topic distributions to determine message relevance. This substitution enables more accurate matching of user interests with message content, improving reliability while reducing the time users spend filtering irrelevant messages.
3Measurement precision
If topic model with large vocabulary is used, then message topic analysis accuracy is improved, but system complexity and computational burden increase
Solution Approach 1:
The patent dynamically adjusts the vocabulary size parameter in the topic model based on computational resources and accuracy requirements. Instead of using a fixed large vocabulary, the system can adaptively select the appropriate vocabulary scale, balancing between topic analysis accuracy and system complexity. This parameter flexibility allows the system to optimize performance according to different operational contexts.
Data Source
AI summary
A method and an apparatus for recommending a message. The method for recommending a message in the present disclosure includes separately parsing a first message published by a first user on a network and a second message published by a second user on the network, obtaining interest description information of the first message and topic description information of the second message, where the second user is another user except the first user, comparing the topic description information with the interest description information, and calculating a similarity of the topic description information and the interest description information; and if the similarity is greater than or equal to a predetermined value, pushing the second message published by the second user to the first user. A user can conveniently and flexibly obtain a message in which the user is interested in the embodiments of the present disclosure.