Conversational Multimedia Recommendation Using Cross-Agent Interaction Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multimedia content recommendation systems fail to accurately reflect user interests beyond historical browsing data, leading to limited and less relevant content suggestions.
Innovation Solution
Integrate an intelligent agent using machine learning models to analyze historical interaction data, including conversation records and agent settings, to determine multimedia content recommendations in a conversation interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If recommendation systems use only historical browsing data, then the system complexity is low, but the recommendation accuracy is insufficient to reflect user interests
Solution Approach 1:
The patent combines multiple data sources (browsing history, conversation records, agent settings) into a unified recommendation system. The determining module integrates these diverse data types to comprehensively analyze user interests, thereby improving recommendation accuracy without creating an unmanageable system structure.
Solution Approach 2:
The recommendation system is designed to handle multiple functions: it processes browsing data, analyzes conversation records, considers agent settings, and generates recommendations. This multi-functional approach allows a single system to reflect diverse user interests while maintaining manageable complexity through unified processing.
2Measurement precision
If the system analyzes multiple types of historical interaction data, then the recommendation accuracy improves, but the data processing complexity increases
Solution Approach 1:
The patent segments the data processing into distinct modules: a determining module that processes browsing history, a separate analysis of conversation records, and consideration of agent settings. This segmentation allows complex multi-type data to be processed systematically and manages processing complexity through structured approach.
Solution Approach 2:
The patent introduces an intermediary processing layer (the determining module and analysis components) that mediates between raw multi-type data and final recommendations. This intermediary structure simplifies the transformation of complex data formats into actionable recommendation insights.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The present disclosure relates to a multimedia content recommendation method and device, an electronic device and a storage medium. The multimedia content recommendation method includes: displaying a first entrance of a recommendation stream of multimedia content in a conversation interface between a user and a first agent; displaying a playing interface of the recommendation stream in response to a trigger operation on the first entrance; determining a multimedia content recommended for the user based on historical interaction data authorized by a user between the user and at least one of the first agent or an agent other than the first agent; and displaying the multimedia content recommended for the user in a playing interface of the recommendation stream.