Browser History Clustering for Topic-Based Search Journeys
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional web browser history pages lack the ability to present search history data in a manner that allows users to assess data from a search topic and journey perspective, with links related to different intents interleaved temporally and no practical way to relate prior searches and page views together around a particular intent, and no mechanism to maintain user privacy and security.
Innovation Solution
Utilize machine learning algorithms to analyze and organize browser history data into cohesive search journeys, generating history clusters that represent search activities related to a particular topic, with user control over data collection and usage, and provide interactive UIs for accessing these clusters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If browser history data is stored sequentially based on access time and date, then the storage structure is simple and easy to implement, but it becomes difficult for users to trace steps back to prior webpage visits and requires extensive manual searching
Solution Approach 1:
The patent segments browser history data into distinct categories (search history, download history, cookie history, cached data) and further organizes them into chronological periods (e.g., last 7 days, last 30 days). This segmentation allows users to navigate through history data more efficiently by focusing on relevant time periods and types rather than searching through all sequential entries.
Solution Approach 2:
The system performs preliminary organization of history data by automatically categorizing and time-grouping entries before the user needs to access them. The browser maintains pre-organized historical data structures that group related activities together, so when a user requests history information, it is already structured and ready for efficient retrieval without requiring manual sorting or searching.
2Measurement precision
If manual searching of stored navigation information is performed, then complete search results can be obtained, but it consumes significant computational resources and battery capacity on mobile devices
Solution Approach 1:
The system performs preliminary filtering and organization of history data by pre-categorizing entries into time periods and types. This preliminary action reduces the amount of data that needs to be searched when a user queries for specific information, thereby reducing computational resources and energy consumption while maintaining complete and accurate search results within the filtered scope.
Solution Approach 2:
By segmenting history data into time-based categories (e.g., recent history, older history) and type-based categories (search, download, cookies), the system enables users to target their searches to specific segments rather than scanning through the entire history. This segmentation strategy maintains search completeness within relevant segments while significantly reducing the computational burden of searching through all data.
3Ease of operation
If browser history data is analyzed and organized automatically, then ease of access to search history is improved, but user privacy and security concerns arise
Solution Approach 1:
The browser performs self-service organization of history data using built-in algorithms that automatically categorize and time-group entries without requiring external processing or cloud-based analysis. This self-service approach maintains ease of access to organized history while keeping all data processing local to the device, thereby preserving user privacy and security.
Solution Approach 2:
The system introduces time periods and data categories as intermediary organizing principles that structure history information without exposing raw user data to external systems. These intermediaries (time ranges, type categories) serve as safe abstraction layers that enable organized access to history while maintaining data privacy and security through local processing only.
Data Source
AI summary
Systems and methods are described that include generating a repository of metadata based on a plurality of webpages accessed and saved in a browser history of a web browser executing on a computing device, generating, based on the metadata, a history cluster including a portion of the plurality of webpages related to a topic where the history cluster generation is based on the source events and the access timestamps of the webpages in the portion, and assigning respective scores for the webpages in the portion. In response to a request to view browser activity associated with the topic, the systems and method may generate and display a history cluster listing for the topic where the history cluster listing includes visit listings associated with the webpages in the history cluster that are determined to have a score that meets a threshold score.