Notification Contextual Information System Using Entity Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing interaction models require users to provide queries to obtain contextual information, which can be time-consuming and prone to errors, especially when users are unable to type or speak.
Innovation Solution
A system that generates query-independent requests for contextual information based on notifications, identifying relevant search items and providing user interface elements with contextual information without user input, leveraging a client contextual module, item identification engine, relevance scoring engine, and contextual card provider.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If users provide queries to obtain contextual information, then the information can be retrieved, but the process becomes time-consuming and error-prone
Solution Approach 1:
The notification system automatically generates queries and retrieves contextual information without requiring user input. The system serves itself by extracting entities from notifications, formulating search queries, and presenting relevant information automatically, eliminating the need for users to manually type or speak queries.
Solution Approach 2:
The system performs information retrieval in advance by automatically processing notifications as they arrive. Contextual information is fetched and prepared before the user needs it, so when the user views the notification, the supplementary information is already available without requiring additional time from the user.
2Loss of information
If users provide queries to obtain contextual information, then the information can be retrieved, but input errors occur when users are unable to type or speak
Solution Approach 1:
The system eliminates the need for user input by automatically extracting entities from notifications and generating search queries. This self-service approach removes the source of input errors entirely, as the system uses the notification text itself as the basis for information retrieval without requiring user typing or speech.
Solution Approach 2:
The notification text serves as an intermediary between the user's information need and the search system. Instead of the user directly inputting queries (which can be error-prone), the notification content mediates the process by providing the search terms automatically, thereby improving reliability.
3Loss of information
If the system processes user queries, then contextual information can be provided, but processing resources are consumed
Solution Approach 1:
The system performs partial information retrieval by focusing only on the most relevant entities extracted from notifications rather than processing complete user queries. By selecting only key entities from notifications and retrieving information specifically about those entities, the system reduces processing resources while still providing sufficient contextual information.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing contextual information for a notification. In one aspect, a method includes receiving, from a user device in response to the user device obtaining a notification in response to an occurrence of a trigger event, a query-independent request for contextual information relevant to the notification, identifying multiple search items from content in the notification, determining a relevance score for each of the multiple search items, selecting one or more of the multiple search items based on the relevance scores, and providing, to the user device for each of the selected one or more multiple search items, a respective user interface element for display with the notification, wherein each user interface element includes contextual information regarding the respective search item.