Soft Skill Inference from Course Syllabi via NLP
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Organizations and educational institutions face challenges in determining interpersonal or behavioral skills from course information, as conventional methods rely on anecdotal evidence and lack evidence-based analytics, making it difficult to assess the skills learned from courses effectively.
Innovation Solution
A system employing skills models and natural language processing to infer soft skills from course information by analyzing learning objectives and syllabi, updating course profiles with inferred skills, and using machine learning to improve skill model accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods are used to determine interpersonal or behavioral skills, then organizations can make hiring decisions, but the accuracy and reliability of skill assessment is poor due to reliance on anecdotal evidence and personal preferences
Solution Approach 1:
The patent replaces manual, subjective assessment methods (anecdotal evidence, personal preferences, gut feelings) with an automated natural language processing system that extracts and analyzes skills from course information. This substitution transforms the assessment from a mechanical human judgment process to an algorithmic text analysis process, improving both precision and reliability through consistent, data-driven evaluation.
Solution Approach 2:
The system enables course information to self-describe its skill content through automated NLP extraction. Instead of requiring manual annotation or expert review of each course, the system automatically processes course descriptions, learning objectives, and syllabi to identify and categorize interpersonal and behavioral skills, allowing the data to serve itself without extensive human intervention.
2Adaptability or versatility
If educational institutions design course offerings based on perception of employer needs, then they can provide relevant education, but the alignment with actual employer requirements is insufficient without evidence-based analytics
Solution Approach 1:
The patent establishes a feedback loop where employer skill requirements are systematically extracted from job descriptions and fed back into course design. The NLP system continuously analyzes employer needs and provides data-driven insights to educational institutions, enabling them to adapt course offerings based on actual market demands rather than perceptions, thus improving alignment and relevance.
3Loss of information
If students evaluate courses based on available course information, then they can make enrollment decisions, but the completeness of skill understanding is limited
Solution Approach 1:
The patent segments course information into distinct skill categories, separating interpersonal and behavioral skills from technical content. By breaking down the course description into structured skill components, the system enables students to clearly see which specific soft skills will be developed, providing a complete and precise view of skill outcomes rather than a generic course description.
4Measurement precision
If employers review course information to evaluate candidate fit, then they can make hiring decisions, but the ability to discern complete skill views from courses is difficult
Solution Approach 1:
The patent introduces an intermediary NLP system that acts as a bridge between course information and employer needs. This intermediary automatically extracts, standardizes, and presents skill information in a format that directly matches employer requirements, eliminating the complexity of manual analysis while providing a complete and accurate skill view for hiring decisions.
Data Source
AI summary
Embodiments are directed to determining interpersonal or behavioral skills based on course information in course offerings. One or more learning objectives may be determined from course information associated with a course such that each learning objective may be associated with a learning objective narrative. A skills model may be employed to determine one or more soft skills based on the one or more learning objective narratives. The one or more soft skills may be associated with a course profile that corresponds to the course. In response to a quality score associated with the skills model being less than a threshold value, the skills model may be retrained with one or more reference models having one or more portions based on machine learning.