Adaptive Message Filtering via Dynamic Concept Interest Datasets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of Twitter-like micro-blog social media services and general message broadcast services face difficulty in finding relevant messages due to the overwhelming volume of irrelevant information, and existing technologies do not adapt to changes in user interests over time.
Innovation Solution
A method and system that identifies user actions on messages, extracts concepts, determines interest values, and updates a concept interest dataset to filter messages based on adaptive thresholds, using a processor to generate a message filter and provide a filtered messaging service that evolves with user interests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If message filtering is implemented to reduce irrelevant information, then message relevance is improved, but system complexity increases
Solution Approach 1:
The system dynamically adjusts filtering parameters including interest value thresholds and concept weights based on user behavior patterns. The threshold for message inclusion is not fixed but adapts as the system learns user preferences, allowing the filter to become more precise without requiring proportional increases in system complexity
Solution Approach 2:
The filtering system automatically updates its own parameters by monitoring user interactions with messages. Through self-service learning mechanisms, the system refines its concept interest dataset and adjusts thresholds without external intervention, achieving improved relevance while minimizing the operational overhead
2Device complexity
If static filtering rules are used to simplify the system, then system complexity is reduced, but adaptability to changing user interests deteriorates
Solution Approach 1:
The system transitions from static filtering rules to dynamic adaptive filtering. The concept interest dataset and threshold values are continuously updated based on observed user actions such as message engagement and interaction patterns. This dynamic adjustment allows the system to adapt to evolving user interests while maintaining manageable complexity through automated learning processes
Solution Approach 2:
The system implements feedback loops where user interactions with filtered messages are monitored and fed back into the filtering algorithm. This feedback mechanism enables the system to learn from user behavior and automatically adjust its filtering criteria, achieving adaptability without requiring complex manual configuration or rule updates
Data Source
AI summary
A method and system. A first interest value is determined. The first interest value is associated with a first combination of one or more concepts derived from one or more word objects extracted from a message on which an action is to be performed. The first interest value is determined to be at least a specified first threshold value and in response, a concept interest dataset is updated. The concept interest dataset prior to being updated includes combinations of at least one concept. Each concept of the at least one concept has been derived from a previous processing of messages. The updated concept interest dataset includes the first combination and a cumulative interest value that encompasses the first interest value.