App Store Search Query Classification and Result Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing number of software applications in app stores makes it difficult for users to find relevant apps, as existing search systems often return overwhelming and irrelevant results, even when using keyword searches.
Innovation Solution
A method that classifies search queries as either navigational or functional, using distribution data to identify similar applications by analyzing download patterns, and sorts search results to prioritize the original application and functionally similar apps, while also incorporating textual matches from application metadata.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If keyword search is used to find apps in a large app store, then the search can cover many applications, but the search results become overwhelming and may not be relevant to the user's need
Solution Approach 1:
The patent segments search queries into two distinct types: navigational queries (when users search for a specific app by name) and functional queries (when users search for apps by function or capability). This segmentation allows the system to apply different search strategies and result presentations for each query type, improving overall search effectiveness by tailoring the approach to the user's specific intent
Solution Approach 2:
The patent dynamically adjusts the search result presentation based on the classified query type. For navigational queries, the system prioritizes exact name matches and provides a focused list. For functional queries, the system provides a broader list sorted by relevance to the functional need. This dynamic adaptation resolves the contradiction by making the search system flexible enough to handle both broad coverage and precise targeting
2Reliability
If the search returns many apps to ensure comprehensive results, then the search is thorough, but the user has difficulty finding the specific app needed
Solution Approach 1:
The patent performs preliminary classification of the search query before executing the full search. By determining whether the query is navigational or functional in advance, the system can prepare and present results in the most efficient format from the start, avoiding the need for users to sift through irrelevant results and reducing the time to find the desired app
Solution Approach 2:
The patent applies different quality standards and sorting criteria to different portions of the search results based on query type. For navigational queries, the top results are optimized for exact name matching. For functional queries, the results are optimized for functional relevance. This local optimization ensures that users immediately see the most relevant results without wasting time reviewing less relevant options
Data Source
AI summary
Methods and systems for searching for applications in an app store use, in one embodiment, distribution data to create an association between a first application and a first set of other applications in the app store, and the association is used to provide search results in response to a navigational search query that specifies the first application by name so that the search results include the first application and the first set of other applications. The distribution data indicates the number of downloads for applications in the app store that resulted from each search query in a set of search queries.