Descriptive Phrase Association Engine for Entity Tagging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search technologies fail to effectively retrieve relevant results when users query for products based on descriptive attributes rather than exact names, leading to missed relevant products in product catalogs.
Innovation Solution
A descriptive phrase association engine that collects and analyzes description indicators and entity indicators from documents to generate description tag associations, enabling more effective search by automatically associating descriptive phrases with entities, even if those phrases are not mentioned in the entity's name or description.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If users search using descriptive attributes rather than exact product names, then search relevance improves, but the ability to retrieve accurate results deteriorates due to lack of precise matching
Solution Approach 1:
The system pre-computes and stores descriptive phrases (etags) for each entity in advance by analyzing web documents. When a search query arrives, the system retrieves entities whose pre-computed etags match the query terms, enabling fast and accurate attribute-based search without real-time analysis overhead.
Solution Approach 2:
The patent introduces descriptive phrases (etags) as intermediary terms between user queries and entity names. These etags act as mediators that bridge the gap between attribute-based user intent and exact entity identification, allowing searches like 'water-resistant camera' to match entities even when those exact terms don't appear in the entity name or description.
2Productivity
If the system analyzes web documents to extract descriptive phrases, then search coverage improves, but processing time and computational resources increase
Solution Approach 1:
The system performs document analysis and etag extraction in advance during an offline phase. The extracted etags are stored in a database associated with their respective entities. During online search operations, the system only performs fast database lookups based on query terms matching pre-computed etags, eliminating the need for real-time document analysis.
3Productivity
If the system associates descriptive phrases with entities based on web document analysis, then recall improves, but precision may deteriorate due to noisy or irrelevant associations
Solution Approach 1:
The system incorporates feedback mechanisms including: (1) requiring multiple document occurrences to establish an etag-entity association, (2) using co-occurrence frequency analysis to weight associations, and (3) allowing manual review and correction of automatically generated etags. This feedback loop filters out noisy associations while preserving relevant ones.
Data Source
AI summary
A plurality of description phrases associated with a first domain may be determined, based on an analysis of a first plurality of documents to determine co-occurrences of the description phrases with one or more name labels associated with the first domain. An entity associated with the first domain may be obtained. An analysis of a second plurality of documents may be initiated to identify co-occurrences of mentions of the obtained entity and one or more of the plurality of description phrases, and contexts associated with each of the co-occurrences of the mentions and description phrases, in each one of the second plurality of documents. A description tag association between the obtained entity and one of the description phrases may be determined, based on an analysis of the identified contexts.