Context Pattern Detection for Term Disambiguation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current natural language processing systems lack an effective means to disambiguate the meaning of terms with multiple meanings across different linguistic domains, leading to inefficiencies and increased costs in applications like drug safety approval and social media sentiment analysis.
Innovation Solution
A method that involves obtaining corpora of words associated with different linguistic domains, generating sets of patterns for each domain, and examining a word of interest in a passage to notify users of its intended sense by identifying matching patterns, using a language-independent approach.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a natural language processing system is expanded to process terms across multiple linguistic domains, then the system's versatility improves, but its ability to accurately disambiguate term meanings deteriorates
Solution Approach 1:
The patent segments the linguistic domain into multiple specialized corpora, each representing a distinct domain (e.g., medicine, law, technology). By dividing the broad linguistic space into manageable segments, the system can maintain high precision within each domain while achieving versatility across domains through the collection as a whole.
Solution Approach 2:
The patent introduces context patterns as an intermediary mechanism that mediates between the ambiguous term and its correct meaning. These patterns serve as domain-specific contextual frameworks that help disambiguate terms by matching them against appropriate domain corpora, resolving the conflict between versatility and precision.
2Measurement precision
If context pattern detection is used to disambiguate terms, then measurement precision of term meaning improves, but device complexity increases
Solution Approach 1:
The patent performs preliminary action by pre-generating context patterns from domain-specific corpora before actual term disambiguation is needed. These patterns are stored and organized in advance, allowing the system to quickly match terms against pre-processed contextual frameworks rather than generating patterns in real-time, thus reducing operational complexity.
Solution Approach 2:
The patent creates simplified copies of domain-specific contextual information in the form of context patterns. Instead of processing entire corpora during operation, the system uses condensed pattern representations that capture essential domain characteristics, reducing computational complexity while maintaining disambiguation accuracy.
3Measurement precision
If multiple domain-specific corpora are maintained, then measurement precision of term disambiguation improves, but loss of information increases due to data management complexity
Solution Approach 1:
The patent extracts essential contextual information from comprehensive domain corpora to create focused context patterns. By taking out only the relevant contextual features needed for disambiguation and storing them as patterns, the system maintains high precision while reducing data management complexity and preventing information loss through efficient representation.
Data Source
AI summary
A method, system and computer program product for disambiguating meaning of terms. A natural language processing system obtains a first corpora of words used in a first sense and a second corpora of the same words that are used in a second sense. Each of these corpora of words may be associated with different linguistic domains. The natural language processing system generates a first and a second set of patterns using both the first and second corpora of words, respectively. A question passage is then received by the natural language processing system. The natural language processing system examines a word of interest in the question passage. The user is then notified that the word of interest is being used in the first sense or the second sense in response to identifying the word of interest in only the first set of patterns or the second set of patterns, respectively.