Indexing Native Application Data via Virtual Machine Copying
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Search engines lack accurate relevance measures for native application content, relying on external metadata which can lead to descriptive errors and requiring content publishers to perform optimizations, and they do not effectively utilize user activity data from native applications.
Innovation Solution
A method where a search engine accesses and indexes data directly from native applications on mobile devices, generating search results that include links to viewed content, improving relevance and reducing errors by using actual application page data rather than metadata, and providing users control over personal information usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If search engines use external metadata to describe native applications, then they can index application information, but the metadata may contain descriptive errors and require publisher optimizations
Solution Approach 1:
The search engine creates a virtual machine that copies and executes the native application to extract data directly from its internal structures, replacing unreliable external metadata with accurate in-application data
Solution Approach 2:
The search engine introduces a virtual machine as an intermediary between itself and the native application, allowing data extraction without direct access to the application's internal memory structures
2Loss of information
If search engines rely on metadata describing native applications, then they can provide search results, but they cannot access actual application page data for comprehensive search
Solution Approach 1:
The search engine creates a virtual machine that copies and executes the native application to extract data directly from its internal structures, replacing unreliable external metadata with accurate in-application data
Solution Approach 2:
The data extraction process is segmented into multiple phases: application execution, data collection from internal structures, and index generation, allowing comprehensive access to application content
3Measurement precision
If search engines index data from native applications through virtual machine execution, then search relevance improves, but system complexity increases
Solution Approach 1:
The search engine creates a virtual machine that copies and executes the native application to extract data directly from its internal structures, replacing unreliable external metadata with accurate in-application data
Solution Approach 2:
The virtual machine automatically executes the native application and extracts data from its internal structures without requiring manual intervention or complex configuration, simplifying the indexing process
Data Source
AI summary
Methods, including computer programs encoded on a computer storage medium, for indexing native application data. In one aspect, a method includes: receiving, by a search engine and from a mobile device, a set of data that is generated by a native application on the mobile device, the set of data including (i) data that identifies the native application, (ii) a representation of viewed content, (iii) a link to the viewed content that, when selected, launches the viewed content in the native application; storing, by the search engine, (i) the data that identifies the native application, and (ii) the link to the viewed content in association with the viewed content; and using the stored set of data in generating a search result that (i) references the viewed content, and (ii) includes the link to the viewed content that, when selected, launches the viewed content in the native application.