A Machine Learning Based Web Platform for Comprehensive Placement Preparation

Abstract

Today's competitive job market expects candidates to demonstrate their abilities through technical skills, analytical skills and strong communication skills as well as having an appropriate resume that reflects the skills needed in the workplace. Unfortunately, many students do not have sufficient access to an integrated system of personalized, centralized and smart preparation tools causing significant barriers to student employability. In light of these deficiencies, this project presents a new approach to a machine learning based web platform that provides students with an all-in-one preparation system that combines curriculum specific content, aptitude practice, resume assessment and a chat bot. The new system employs machine learning and natural language processing techniques including supervised classification models, keyword extraction, semantic similarity assessment and adaptive assessment algorithms to create a customized preparation process for each student. A combination of back-end application programming interfaces will be used to provide machine learning based performance evaluation and actionable feedback as well as real-time analytics and predictive models to enable users to identify their level of readiness for placement and to identify gaps or weaknesses in their skill sets needing improvement. Results from the evaluation of this system indicate the machine learning based platform will dramatically assist students in preparing for their placement by improving scores from quizzes taken during preparation, improving quality of resumes being submitted, and increasing the level of confidence of students moving into their placement process. Overall this project demonstrates that machine learning based platforms can reinvent how training and preparation occur for placements by providing institutions and students with scalable, efficient and data driven preparation solutions.