Context-Aware App List Provisioning via Message Keyword Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face inconvenience in finding relevant apps during conversations on portable devices due to the need for manual app searches, especially on wearable devices with limited display space, where providing a context-fitting app list based on message text and context is lacking.
Innovation Solution
A method and device that analyze messages to determine keywords, generate deep links for relevant apps, and provide a customized app list, including installation links for non-installed apps, using named-entity recognition and personalized information to enhance user convenience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If users manually search for apps during conversations on portable devices, then they can find relevant apps, but user convenience deteriorates due to the manual effort required
Solution Approach 1:
The system performs preliminary analysis of message text and context to identify relevant apps before the user needs them. By pre-processing the conversation content and matching it with installed applications, the system prepares app recommendations in advance, eliminating the need for manual searching and improving user convenience.
Solution Approach 2:
The system automatically provides app recommendations based on the conversation context without requiring user intervention. The messaging application autonomously analyzes the message, determines relevant apps, and presents them to the user, allowing the system to serve itself rather than requiring manual app searching.
2Area of stationary object
If wearable devices are designed with traditional size and shape constraints, then user comfort is maintained, but display area is insufficient for showing app lists
Solution Approach 1:
The system segments the app list display into multiple pages or sections that can be navigated sequentially on the limited wearable device display. Instead of attempting to show all apps simultaneously, the interface divides the content into manageable portions, allowing comprehensive information presentation without requiring a larger display area.
Solution Approach 2:
The system transitions from a two-dimensional display limitation to utilizing temporal navigation through multiple display pages. By adding the time dimension for sequential viewing, the limited display area can present comprehensive app information through pagination, scrolling, or category-based grouping that unfolds over time rather than requiring expanded spatial area.
3Loss of information
If the app list is customized per user and context, then relevance is improved, but system complexity increases due to multiple analysis parameters
Solution Approach 1:
The system implements a universal context analysis framework that handles multiple parameters (user profile, message text, conversation context, time, location) through a single integrated process. This multi-functional approach allows the same core algorithm to process various types of information and generate customized app recommendations without requiring separate complex systems for each parameter.
Solution Approach 2:
The system dynamically adjusts the weight and importance of different context parameters based on the situation. By changing the parameters of the analysis process itself (such as emphasizing certain keywords or user preferences in specific contexts), the system achieves high customization without requiring equally complex processing for all possible scenarios, optimizing the balance between relevance and complexity.
Data Source
AI summary
Provided is a method for providing a related application list on the basis of a message. An application list providing method according to an embodiment of the present invention, which is performed by an application list providing device, may comprise the steps of: receiving a provided message; determining a keyword on the basis of the message; determining an application list associated with the message by using the determined keyword, wherein the application list includes at least one application (APP); generating a deep link for at least one application included in the application list by using the determined keyword; and providing the application list and the deep link.