Application Recommendation Segmentation for Search Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current electronic devices face challenges in providing an efficient user search experience for applications, as users often need to repeatedly search for related applications, leading to a poor recommendation effect.
Innovation Solution
An application recommendation method that clusters applications based on multiple tag types, allowing users to view applications hierarchically and compare different types, thereby improving the user retrieval experience and increasing the conversion rate of application recommendations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If applications are displayed in a tiled manner after search, then all related applications can be shown, but users need to repeatedly search and cannot quickly find specific applications
Solution Approach 1:
The patent segments applications into different categories (e.g., recommended applications, popular applications, new applications) and displays them in distinct sections on the search results page. This segmentation allows users to quickly locate specific types of applications without searching through all results, thereby reducing search time while maintaining comprehensive information display.
Solution Approach 2:
The patent introduces a categorical dimension to the display by organizing applications into multiple classified sections rather than a single flat list. This dimensional organization enables users to navigate through different application types systematically, improving search efficiency without losing any application information.
2Adaptability or versatility
If multiple types of applications are displayed together, then users can compare different applications, but the interface becomes complex and hard to navigate
Solution Approach 1:
The patent divides the interface into clearly separated sections for different application types (recommended, popular, new applications). Each section has a distinct header and organized layout, making the complex information structure easier to navigate while preserving the ability to compare multiple applications across different categories.
3Measurement precision
If applications are recommended based on search content, then recommendation correlation improves, but the recommendation system becomes complex
Solution Approach 1:
The patent pre-categorizes applications into multiple types (recommended, popular, new) before displaying them on the search results page. This preliminary organization based on search content allows the system to provide correlated recommendations without requiring complex real-time processing, as the categorization logic is prepared in advance.
Data Source
AI summary
An application recommendation method includes: receiving search content input on a first interface of an application market application; and displaying a second interface of the application market application in response to the search content, where the second interface includes a first tag type, a second tag type, first application information, and second application information, the first application information includes a name and an icon of a first application, the second application information includes a name and an icon of a second application, a tag type of the first application is the first tag type, a tag type of the second application is the second tag type, and the first tag type is different from the second tag type.