ML based career suggestive system for informal job sector considering cognitive skills

Abstract

We are striving to build a realistic procedure by which we, particularly the future generation, will be able to choose the right career based on their capacity and interests. A few well-known international firms, including IBM, Unilever, LinkedIn, Accenture, and others, utilize Pymetrics to hire their staff, which is based on cognitive skills in the formal sector. Our work, however, is the first in the informal sector. On our primary collected dataset, we used six distinct algorithms, including Logistic Regression, Decision Tree, Random Forest Classifier, Support Vector Classification, Multilayer Perceptron Classifier, and Extreme Gradient Boosting (XGB), and discovered that Random Forest Classifier and Extreme Gradient Boosting (XGB) are the best for this system, with the accuracy of 57% and 60%, respectively. We’ve also used MinMaxScaler to enhance our output. After that, we observed that the Random Forest Classifier approach had a nearly 62% higher accuracy. The Extreme Gradient Boosting (XGB) approach, on the other hand, has a precision of 58.6%. After completing our evaluation, we opted to use the Random Forest Classifier for our system instead of MinMaxScaler. Based on these insights, we’ll match individuals with employment, smoothing out labor market inefficiencies and leading to considerable boosts in productivity, income, and well-being.