Semantic Indexing via Frame Mapping for Focused Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computerized search engines, both keyword and semantic, often provide unfocused results due to the large number of semantic meanings and patterns associated with sentences, leading to missed desired outcomes.
Innovation Solution
A semantic indexing system that maps textual documents to frames carrying semantic meaning, allowing for focused searching by generating an index that maps statements to frames, with options for resolving incompatibilities through user input and updating the index.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If semantic search engines map sentences to multiple semantic patterns to understand meaning, then the semantic understanding capability is improved, but the number of results becomes unmanageably large and unfocused
Solution Approach 1:
The patent segments the semantic search process into two distinct phases: (1) an offline indexing phase where sentences are mapped to semantic patterns and stored in a pre-computed index, and (2) an online query phase where the pre-built index is used for efficient retrieval. This segmentation allows the system to handle complex semantic mappings without overwhelming the search process, as the heavy lifting of pattern matching is done beforehand and stored for quick access.
Solution Approach 2:
The patent performs preliminary action by pre-computing and storing the mapping between sentences and semantic patterns in an index before actual search queries are executed. This offline preparation includes identifying semantic patterns, creating the mapping relationships, and storing them in a structured format. When a search query arrives, the system can quickly retrieve pre-computed results rather than performing complex semantic analysis in real-time, thus managing the large number of patterns efficiently.
2Speed
If keyword search engines match keywords to text to perform search, then the search speed is improved, but the results miss the semantic meaning and desired outcomes
Solution Approach 1:
The patent introduces semantic patterns as an intermediary layer between keyword matching and semantic understanding. Instead of directly matching keywords to text or performing complex full semantic analysis, the system maps both queries and documents to intermediate semantic patterns. This intermediary representation captures the essence of semantic meaning while maintaining computational efficiency, allowing the system to retrieve semantically relevant results without the full overhead of deep semantic analysis for every query.
3Adaptability or versatility
If the number of semantic patterns is increased to cover more meanings, then the coverage of search results is improved, but the precision and focus of results deteriorates
Solution Approach 1:
The patent implements dynamic filtering and ranking mechanisms that adapt to each specific query context. Rather than using a static threshold for all searches, the system dynamically adjusts which semantic patterns are applied and how results are ranked based on the query's specific requirements. This dynamic approach allows the system to maintain high coverage by considering multiple semantic patterns while preserving precision through context-aware filtering and prioritization of the most relevant matches.
Data Source
AI summary
A semantic indexing system, the semantic indexing system comprising a processing resource configured to: provide a corpus comprising a plurality of textual documents, wherein (a) each of the textual documents being composed of one or more sentences; (b) each of the sentences being composed of one or more statements; and generate an index, the index mapping each of the statements to one or more frames; wherein each frame defines a structure that carries a semantic meaning, thereby enabling searching the corpus by the semantic meaning of a search statement.