Journal of Artificial Intelligence and Computational Technology

Resume Screening & Job Matching System: A Machine Learning Approach

S B Anusha — 2025-11-15

A Machine Learning and Natural Language Processing (NLP) Powered Resume Screening and Job Matching System is an indispensable tool in current hiring. Manual screening of hundreds of resumes takes too much time and is liable to bias, especially in this highly competitive era of job-hunting. Companies can enhance productivity, decrease the time-to-hire, and make more impartial hiring decisions with the automation process. The system would scan resumes, extract pertinent information like qualifications, skills, experience, and certifications, and cross-check them against job descriptions for compatibility scores. This would help recruiters immediately filter out the most appropriate candidates without manually sifting through each resume. This will give rise to an efficient system of matching the requirements of an employer with the ability to do so by an individual in the present highly competitive job market. Resume Screening and Job Matching System-Applying machine learning with natural language processing-an Automated Candidate Selection Process. This allows the system to parse, analyze, and bring up the core qualifications, skills, and experiences contained within the resumes and to find alignments with the descriptions in a posted job, thereby yielding compatibility scores. These aim at improving hiring efficiency and reduce time-to-fill along with minimizing bias. Through the use of NLP, the system extracts and processes the pertinent features like skills, experience, qualifications, and job preferences, allowing for effective comparison of job requirements and candidate profiles. Machine learning classifiers subsequently assess the compatibility of candidates and job roles, generating compatibility scores that enable recruiters to make informed choices. Word embeddings, TF-IDF, and deep learning models like BERT and LSTMs also enhance text comprehension and semantic matching, resulting in more precise recommendations.

Language Translation Application based on Machine learning

Eileen Maria Tom — 2025-11-15

Langage barriers present a significant challenge in education, particularly in diverse classrooms where

students and teachers speak different languages. Artificial Intelligence (AI) offers promising solutions through advanced language translation systems that can deliver fast and accurate translations, enabling students to grasp lessons regardless of language differences. However, current translation tools face limitations, such as difficulties with real-time processing, contextual understanding, and maintaining linguistic nuances. This book chapter delves into the transformative role of AI in education, focusing on the development and deployment of machine learning-powered translation technologies. It examines how integrating AI-driven solutions can enhance learning experiences, promote inclusivity, and overcome linguistic obstacles. By improving translation accuracy and adaptability, educational systems can become more accessible and equitable, fostering a global learning environment that supports students from diverse cultural and linguistic backgrounds.

Designing A Model to Analyze the Impact of Applied and Mathematical Subjects on Student Academic Achievement

Babiker Gsmallah ELshiekh Mustafa — 2025-11-15

Educational institutions need to analyze their data to improve the educational process,
which faces numerous challenges, such as the difficulty of measuring learning outcomes and the
factors that influence them, and the lack of knowledge about the causes of student academic
decline. This research aims to provide a comprehensive study of the use of data mining techniques
in educational institution data. This study utilizes classification technology, one of the most
important data mining techniques, and applies it to a sample of students from the College of
Computer Studies and Information Technology at the Universities of Omdurman Islamic University and West Kordofan University.

The data were analyzed using their grades. Five classification models were constructed and
compared in terms of the accuracy of the results, model construction time, and error rate. The goal
was to ensure the best results. The JRIP algorithm was chosen, achieving the best results among
the algorithms based on the specified factors. This enabled the researcher to clearly present the
results, analyze, and understand the extent of the impact of practical subjects on student academic
achievement compared to theoretical subjects, and propose appropriate solutions to be presented
to relevant educational authorities to assist them in decision-making.