Application Display Prediction Using Association Rules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for displaying applications on terminals require collecting and processing large amounts of sensor and external data, leading to complex calculations and degraded processing performance.
Innovation Solution
An application display method and apparatus that predicts and displays a second application based on stored sequential or combinational association relationships between applications, established through log information, eliminating the need for extensive data collection by using time series frequent subsequence mining and frequent item mining algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data and external data are collected to predict applications, then application prediction accuracy is improved, but processing performance is degraded
Solution Approach 1:
The patent extracts only the necessary association relationship data from the complex sensor and external environment data. Instead of collecting and processing all sensor data (accelerometer, gyroscope, proximity sensor, etc.) and external data (weather, location, news), the system extracts pre-established application association relationships from usage logs, significantly reducing data volume while maintaining prediction accuracy.
Solution Approach 2:
The patent performs preliminary action by pre-collecting and storing application association relationships during normal usage in the background. The system continuously monitors application usage patterns and builds association models (sequential and combinational) in advance, so that when prediction is needed, only the pre-processed association data needs to be queried, not raw sensor data.
2Adaptability or versatility
If complex calculation is performed on sensor data, then application prediction capability is improved, but computational load increases
Solution Approach 1:
The patent changes the parameters from raw sensor data (acceleration values, temperature, humidity, GPS coordinates) to application association relationships (sequential patterns like A→B, combinational patterns like {A,B,C}). This parameter transformation converts complex numerical calculations into simple pattern matching and probability lookups, reducing computational complexity while preserving prediction capability.
Solution Approach 2:
The patent creates a simplified copy of the usage environment by storing application association relationships that represent the essential patterns of user behavior. Instead of replicating the entire sensor data processing pipeline, the system uses pre-mined association rules (using algorithms like Apriori and SPADE) that capture the core predictive information in a compact form.
Data Source
AI summary
Embodiments of the present invention provide an application display method and apparatus. A second application that has an association relationship with an application (a first application) that is currently displayed on a terminal is acquired according to a stored association relationship between applications of the terminal, and the second application is displayed after the first application exits. That is, an application that a user expects to use is predicted according to the stored association relationship between the applications, and a large amount of sensor data of the terminal and external data do not need to be collected, thereby improving processing performance of the terminal.