Cognitive profiling system for enhancing placement opportunities in Higher Education

Abstract

The goal of the study is to address the difficulty of predicting engineering students' placement success through the use of cognitive and psychological assessment data, such as Multiple Intelligence, Big 5 Personality traits, Learning Styles, Thinking Styles, and Metacognitive Awareness. By using psychological profiles to predict employability outcomes, this novel method assists recruiters and educational institutions in determining students' placement readiness. After preprocessing, the data from 87 of the 102 students who took the survey were examined. Every stage of the placement evaluation—the Quantitative Aptitude Test (QAT), Verbal and Non-Verbal Reasoning (VNR), Computer Proficiency Test (CPT), Programming and Coding Test (PCT), English Communication Test (ECT), Technical Round (TR), and HR Round—has key cognitive factors that have been identified. The mean values of these cognitive factors were used as inputs for the Mean Difference-Based Clustering and Multiple Criteria Decision-Making (TOPSIS) algorithms, resulting in visualizations of learner categorization, attribute ranking and prioritization, and the identification of the optimal placement solution.

Specification

Description:Detailed Description of the Invention
1. Data Collection and Participant Selection:
Participants: Select a group of engineering students (e.g., 102 students) from a higher education institution.
Data Collection: Collect data on psychological and cognitive attributes such as:
o Multiple Intelligences: Assessing individual strengths in areas like logical, linguistic, spatial, etc.
o Big 5 Personality Traits: Evaluating personality dimensions like openness, conscientiousness, extraversion, agreeableness, and emotional stability.
o Learning Styles: Identifying how students prefer to learn (visual, auditory, kinesthetic, etc.).
o Thinking Styles: Understanding whether students lean towards creative, analytical, or practical thinking.
o Metacognitive Awareness: Evaluating students' awareness of their cognitive processes and learning strategies.
2. Preprocessing of Data:
Data Cleaning: Clean the collected data to remove inconsistencies, missing values, and outliers. After preprocessing, 87 students' data will be considered for further analysis.
Normalization/Standardization: Standardize or normalize the data to ensure all attributes are on a comparable scale.
Handling Missing Data: Apply imputation techniques for missing values or remove records with incomplete data.
3. Cognitive Profiling:
For each student, assess and profile based on the following cognitive aspects:
o Quantitative Aptitude Test (QAT): Identify cognitive traits that influence problem-solving and numerical ability.
o Verbal and Non-Verbal Reasoning (VNR): Assess cognitive aspects related to logic, reasoning, and pattern recognition.
o Computer Proficiency Test (CPT): Evaluate knowledge and practical skills in computer technologies.
o Programming and Coding Test (PCT): Measure technical and coding abilities.
o English Communication Test (ECT): Gauge English language proficiency in written and verbal communication.
o Technical Round (TR) and HR Round (HR): Identify social and technical traits crucial for placement rounds.
4. Mean Difference-Based Clustering:
Clustering Model: Implement the Mean Difference-Based Clustering algorithm to classify students based on their cognitive scores.

o Input Variables: Use the mean scores from various cognitive dimensions (QAT, VNR, CPT, PCT, ECT, TR, HR).
o Cluster Formation: Group students into categories based on their cognitive profiles to identify learners with similar strengths and weaknesses.
5. Multi-Criteria Decision Making (TOPSIS):
Apply TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) to rank and prioritize attributes for placement readiness.
o Criteria: Use cognitive and psychological aspects as criteria for ranking.
o Objective: Determine which cognitive attributes are most important for achieving successful placement outcomes.
o Best Ideal Solution: Identify the ideal cognitive profile for optimal employability based on the clustering and ranking results.
6. Learner Categorization:
Visualization of Categorization: Generate visual outputs, such as graphs and heatmaps, to show learner distribution in four categories based on their cognitive profiles:
o Excellent: High scores in both QAT and VNR.
o Good: High QAT, low VNR.
o Average: Low QAT, high VNR.
o Poor: Low scores in both QAT and VNR.
Categorization Outcome: Provide insights into students' placement readiness and identify students who need support in specific cognitive areas.
Attribute Ranking and Prioritization:
Solving MCDM problem using TOPSIS Method , Claims:1. A method for analyzing and categorizing learner characteristics comprising:
.Receiving input data related to multiple intelligences, learning styles, thinking styles, metacognitive awareness, and personality traits;  Preprocessing the input data to prepare it for analysis;  Normalizing the preprocessed data to standardize its values;  Categorizing the normalized data into defined groups based on predetermined attributes; .Applying mean difference-based clustering to categorize learners into groups with similar characteristics;  Utilizing a multi-criteria decision-making algorithm, specifically TOPSIS, to rank the categorized attributes and determine an ideal solution;  Generating visualizations of learner categorization based on clustering results;  Providing an attribute ranking and an ideal solution to assist in educational interventions or personalized learning paths.
12 | P a g
e
2. The method of Claim 1, wherein the data preprocessing step includes cleaning, encoding, and transforming the input data to improve quality and compatibility for analysis. 3. The method of Claim 1, wherein the normalization step involves adjusting values to fit within a predefined range, enhancing comparability across various learner attributes. 4. The method of Claim 1, wherein the categorization into groups includes classification into specific attributes such as QAT, VNR, CPT, PCT, TR, ECT, and HR, each representing unique characteristics of learners. 5. The method of Claim 1, wherein the mean difference-based clustering is performed by computing the average differences in attribute values between learners, grouping them based on similarity. 6. The method of Claim 1, wherein the multi-criteria decision-making algorithm (TOPSIS) ranks learner attributes by comparing each attribute's proximity to an ideal solution, supporting decisions for learner classification and intervention. 7. The system of Claim 1, wherein the visualization module displays the categorized learners in graphical formats, enabling easier interpretation and decision-making.

Documents