Personalized Search Suggestions via Behavior Graph
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search mechanisms fail to effectively surface relevant people and content items of particular importance or relevance to users, requiring users to manually parse through lists of search results to find relevant information.
Innovation Solution
A personalized search suggestion system that learns from user behavior to provide tailored suggestions for people and content items, using a graph-based approach to aggregate and rank interactions, relationships, and activities, offering two types of suggestions: people and actor-action textual suggestions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional search mechanisms are used to return flat lists of information, then all search results can be displayed, but users cannot efficiently find relevant people or content items of particular importance
Solution Approach 1:
The patent introduces an intermediary suggestion system that sits between the search query and the flat list of results. This intermediary layer analyzes user behavior data, relationships, and interactions to generate personalized suggestions that highlight the most relevant people and content items before the user even sees the complete search results list, thereby reducing the time needed to find important information.
Solution Approach 2:
The system performs preliminary analysis of user behavior, relationships, and interactions in advance to pre-compute and present personalized suggestions. By preparing and displaying relevant suggestions before the user needs to search through all results, the system reduces the cognitive load and time required to identify important people or content items in the search results.
2Productivity
If personalized suggestions are generated based on user behavior, then search efficiency is improved, but the system requires complex data processing and user behavior tracking
Solution Approach 1:
The patent leverages existing multi-functional platforms (such as enterprise social networks or collaboration tools) that already track user behavior, relationships, and interactions for other purposes. By reusing this existing behavioral data infrastructure, the suggestion system avoids the need to build a separate complex tracking system, thereby reducing overall system complexity while still enabling personalized suggestions.
Solution Approach 2:
The system implements feedback loops where user interactions with suggestions are tracked and used to refine future suggestions. This feedback mechanism allows the system to learn and adapt to user preferences over time, improving search efficiency progressively while the complexity increases only incrementally as the system learns from actual usage patterns.
3Loss of information
If flat search results are returned without personalization, then the search mechanism remains simple, but relevant information is not surfaced to users
Solution Approach 1:
The patent applies local quality by providing personalized suggestions tailored to each individual user's context, relationships, and behavior patterns, rather than applying a uniform search result format to all users. This localized personalization ensures that each user sees the most relevant information for their specific needs and relationships, maximizing information relevance without requiring complete system reconfiguration.
Data Source
AI summary
Personalized search or query suggestions associated with one or more persons and/or content items are provided. A suggestion application learns from user behavior within the suggestion application and presents suggestions for allowing the user to search or navigate to one or more people of particular interest or relevance to the user and for allowing the user to search or navigate to one or more content items associated with people and activities of particular interest or relevance to the user. Two types of suggestions are provided to the user. A first type of suggestion involves suggesting one or more people that may be of particular relevance or interest to the querying user. A second type of suggestion includes a textual suggestion comprised of a person (actor) and an associated action.