Personalized Search Summaries Using User Intent and Generative AI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search systems fail to provide personalized search results and summaries that align with the user's intent, requiring users to sift through numerous documents with varying opinions, which is time-consuming and inefficient.
Innovation Solution
A system that analyzes user profile data, intent, and activity history to generate personalized search results and summaries by utilizing a machine-learning model to determine user intent and tailor search results based on the knowledge graph, incorporating generative AI to create concise summaries aligned with the user's current intent.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional search systems present all search results documents to users, then users can access comprehensive information, but users must spend excessive time reviewing numerous documents with varying opinions
Solution Approach 1:
The patent introduces an AI-generated summary as an intermediary between the search query and the full documents. This summary synthesizes key information from multiple documents with varying opinions, allowing users to quickly grasp the essential points without reading every document. The summary acts as a mediator that preserves comprehensive understanding while dramatically reducing review time.
Solution Approach 2:
The system extracts the most essential information from numerous search result documents and presents it in a condensed summary format. By taking out only the critical points, key arguments, and relevant details from multiple sources, the system enables users to access comprehensive understanding without the time cost of reviewing all original documents.
2Productivity
If search systems provide personalized results based on user intent, then search efficiency improves, but system complexity increases due to need for user analysis
Solution Approach 1:
The system employs machine learning models that automatically analyze user profiles, search history, and query patterns to infer user intent without requiring explicit user input or complex manual configuration. This self-service approach to understanding user needs enables personalization while keeping the interface simple and the operational complexity manageable.
Solution Approach 2:
The system dynamically adjusts search parameters and result presentation based on inferred user intent, such as changing the weight of different factors in result ranking or modifying the summary generation approach. By changing parameters automatically based on user characteristics, the system achieves personalization without requiring complex structural changes or manual intervention.
3Loss of time
If search systems generate customized summaries based on user intent, then relevant information is presented quickly, but the complexity of generating accurate summaries increases
Solution Approach 1:
The patent replaces manual information synthesis and summary creation with an AI-based generative system. Instead of requiring human editors to manually create summaries, the system uses machine learning models to automatically generate accurate, intent-aligned summaries from search results, dramatically reducing the complexity of the generation process while maintaining high quality.
Solution Approach 2:
The system performs preliminary analysis of user intent and search results before generating the final summary. By pre-processing the information, identifying key themes, and structuring the content in advance, the system simplifies the final summary generation step and reduces the overall complexity of the process while delivering relevant information quickly.
Data Source
AI summary
Techniques are provided for presenting a summary with search results. One method includes operations for receiving a search query, performing a search based on the search query, and determining, by a machine learning (ML) model, a user intent based on input comprising the search query, user profile information, and activity information of the user. A prompt is generated based on the search query, the user intent, and relationships between the user and entities. The prompt generation comprises selecting a prompt template comprising fields for the user intent, the search query, and the relationships, and filling in the fields. The method further includes operations for feeding as input the prompt to a generative artificial intelligence (GAI) model, receiving a summary from the GAI model, and causing presentation of the summary and one or more results returned by the search on a user interface (UI) of the device of the user.