Intent-Driven Search Result Presentation via Keyword Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines fail to accurately reflect user intent in search results, leading to irrelevant information and inefficient browsing experiences, as they treat queries as mere 'bags of words' without considering the user's specific goals or interests.
Innovation Solution
Implementing a system that analyzes user queries to identify intent through keyword classification, ranking, and presentation rules, which customize search results by assigning intents to data objects and modifying their display to align with user goals, such as product reviews or purchases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If search engines treat queries as bags of words without intent analysis, then the system complexity is low and processing is fast, but the search result relevance to user intent deteriorates
Solution Approach 1:
The patent segments the search process into distinct modules: query analysis module that identifies keywords and intent, search module that retrieves data objects, and ranking module that orders results based on intent matching. This segmentation allows intent-driven search without overwhelming system complexity.
Solution Approach 2:
The patent introduces an intent classification intermediary that acts as a mediator between the user's query and the search results. This intermediary analyzes query keywords to determine user intent and uses this intent information to filter and rank results, improving relevance without requiring complete restructuring of the search system.
2Reliability
If search engines perform comprehensive intent analysis and customization, then the search result relevance improves, but the processing time and complexity increase
Solution Approach 1:
The patent applies preliminary action by pre-defining a set of common user intents and their associated presentation rules before search execution. During search, the system quickly matches query keywords to these pre-defined intents rather than creating custom analysis rules on-the-fly, significantly reducing processing time while maintaining intent-driven relevance.
Solution Approach 2:
The patent changes the parameter of intent classification from a continuous, custom analysis to a discrete selection from pre-defined intent categories. This parameter change allows the system to determine user intent by matching against a finite set of predefined categories, reducing computational complexity and processing time while preserving the ability to deliver relevant results.
3Adaptability or versatility
If search engines assign multiple intents to data objects, then the adaptability to different user goals improves, but the device complexity and processing overhead increase
Solution Approach 1:
The patent implements universality by enabling data objects to be associated with multiple intents simultaneously. A single data object can serve multiple user goals (e.g., a product page can satisfy both purchase intent and review reading intent). This multi-functionality approach increases adaptability without requiring separate processing for each intent, as the system naturally handles multi-intent associations through the ranking mechanism.
Data Source
AI summary
A system and method for intent driven search presentation. A query is received, over a network, from a user, wherein the query comprises at least one query token. The query is analyzed wherein at least one query keyword is identified in the query. The query is classified wherein the query is classified into at least one intent using query keywords. A plurality of data objects that match query keywords is identified. The data objects are ranked, wherein at least one intent is assigned to at least some of the data objects. A result is built using the ranked plurality of data objects, wherein the result comprises display entries wherein if a data object has been assigned at least one intent, such intent is used to construct the display entry for the respective data object. The result is transmitted over the network to the user.