Keyword Selection for Skills Inference Using AI Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for assessing skills and inferring skill expertise levels in organizations are inaccurate, particularly for non-technical skills and technical skills lacking differentiating keywords in the Expertise Taxonomy.
Innovation Solution
A method using trained artificial intelligence models to extract and rank keywords from raw data, incorporating Part-of-Speech tagging and dependency parsing, to improve the accuracy of skill expertise level inference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional keyword-based methods are used for skills assessment, then the process is simple and fast, but the accuracy of skill expertise level inference is low
Solution Approach 1:
The patent segments the keyword extraction and analysis process into multiple distinct stages: (1) extracting keywords from raw text data, (2) determining Part-of-Speech tags for each keyword, (3) parsing dependency relationships between keywords, (4) scoring keywords based on multiple criteria, and (5) selecting final keywords. This segmentation allows each component to be optimized independently while maintaining overall system accuracy.
Solution Approach 2:
The patent applies preliminary actions by pre-processing the raw text data through POS tagging and dependency parsing before keyword extraction and scoring. These preliminary linguistic analyses prepare the data structure in advance, enabling more accurate keyword identification and differentiation of expertise levels without adding complexity during the actual assessment phase.
2Measurement precision
If comprehensive keyword analysis with POS tagging and dependency parsing is implemented, then keyword differentiation accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The patent implements partial action by selectively applying POS tagging and dependency parsing only to keywords that meet certain initial criteria, rather than processing every word in the raw text. The system extracts candidate keywords first, then applies comprehensive linguistic analysis only to these candidates, reducing overall processing time while maintaining accuracy for the most relevant terms.
Solution Approach 2:
The patent changes parameters by introducing multiple scoring dimensions (frequency, distinctiveness, POS type, dependency role) rather than relying on a single keyword match criterion. This multi-parameter approach allows the system to prioritize keywords that provide the most discriminatory power for expertise level differentiation, improving accuracy without uniformly increasing processing complexity across all data.
3Reliability
If multiple scoring criteria (frequency, similarity, distinctiveness) are used for keyword selection, then the quality of curated keywords improves, but the complexity of the selection process increases
Solution Approach 1:
The patent segments the keyword selection process into distinct scoring stages, where each criterion (frequency, similarity, distinctiveness) is calculated and weighted separately. This segmentation allows the system to systematically evaluate keywords against multiple dimensions without creating a monolithic complex process, as each scoring component can be independently optimized and adjusted.
Solution Approach 2:
The patent creates a universal scoring framework that can evaluate keywords across multiple criteria simultaneously. The same keyword candidate is assessed against frequency, similarity, and distinctiveness metrics within a unified selection process, allowing the system to maintain reliability through comprehensive evaluation while avoiding the need for separate complex processes for each criterion.
Data Source
AI summary
A method, computer program product, and computer system for determining keywords from raw data. A first trained artificial intelligence (AI) model is used to extract, from text associated with first and second persons, a first and second plurality of keywords for each first and second person, respectively. A second AI model is used to determine, for each keyword, a similarity score which is zero for each keyword consisting of a single word and is a numerical measure of similarity between the two words in each keyword consisting of two words. A composite score is computed as a function of a keyword frequency rank, a similarity rank, and a person frequency rank for each keyword. A final list of keywords is generated and consists of keywords in the second plurality of keywords based on the composite score of all of the keywords in both the first and second plurality of keywords.