Mobile App Context Extraction via Variable Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in mobile devices is extracting context from diverse and heterogeneous applications, where content access is passive and less predictable, making it difficult to determine user context effectively, especially with growing privacy concerns.
Innovation Solution
A method that retrieves a variable data set from a mobile device's memory, assigns clusters based on data elements, determines the application type, and analyzes content using strategies tailored to each application type, while also considering meta-information and applying privacy rules to ensure user privacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If context extraction is performed from mobile applications, then content understanding and user experience improvement are achieved, but privacy concerns and data protection issues worsen
Solution Approach 1:
The patent extracts only the necessary context information from application views by analyzing variable datasets and their relationships, rather than accessing all user data. The system identifies and extracts specific variables that characterize the view (such as content type, user interaction patterns) while leaving sensitive user information untouched, thus resolving the contradiction between obtaining context and protecting privacy.
Solution Approach 2:
The patent introduces an intermediary analysis device that acts as a mediator between the mobile device and the context extraction process. This intermediary component processes variable datasets in a controlled manner, applying privacy rules and filters to ensure that only appropriate context information is extracted and transmitted, thereby addressing privacy concerns while maintaining effective context understanding.
2Measurement precision
If data extraction is performed from heterogeneous applications, then application type determination is improved, but system complexity increases
Solution Approach 1:
The patent segments the context extraction process into distinct manageable steps: retrieving the variable dataset, analyzing variable types and relationships, assigning clusters to data elements, and determining application type. This segmentation allows the system to handle heterogeneous applications systematically by breaking down the complex task into sequential operations, each dealing with a specific aspect of the data, thereby reducing overall system complexity while maintaining precision.
Solution Approach 2:
The patent employs parameter changes by transforming the raw variable data into clustered categories and application type labels. By changing the representation of data from raw values to semantic clusters (e.g., grouping variables by type and relationship), the system simplifies the analysis process and reduces complexity while improving the precision of application type determination through standardized classification parameters.
3Ease of operation
If passive content access is used in applications, then user experience is improved, but context predictability and user guidance worsen
Solution Approach 1:
The patent implements feedback mechanisms by analyzing user interactions with the application view (such as scrolling patterns, time spent on content, and selection behaviors) and using this feedback to refine context extraction. The system continuously updates its understanding of the user's context based on observed interactions, thereby maintaining predictability even in passive consumption scenarios where users are not actively searching for information.
Data Source
AI summary
This disclosure relates to extracting data associated with a view of an application from a plurality of applications, said application being executed by a mobile computing device and said view being displayed at a display wherein said method comprises the steps of retrieving a variable data set associated with said view of said application from a memory of said mobile computing device, said variable data set comprising at least one data element, said at least one data element of said variable data set characterizes said view of said application and assigning a cluster to each of said at least one data element of said variable data set based on said variable type, said variable with corresponding value and a predefined set of clusters and determining the type of application based on said cluster assigned to each of said at least one data element of said variable data set.