(ICCSAI - 2025)
#64074
2025 3rd International Conference on
Communication, Security, and Artificial Intelligence
Paper ID:- 1130
Title:- Sign language Recognition Using Deep Learning
Track:- 14
Presenter Name:- Aditya Raj , Nishi Kant Gupta
Designation:- Student
Affiliation:- SCSE, Galgotias University
Outline
• Introduction
• Existing Approaches/Related Works
• Problems in Existing Approaches
• Proposed Methodology
• Results and Discussion
• Conclusions and Future Work
• References
Introduction
Resume Analysis Through Machine Learning
Hiring for jobs comes with many challenges like mistakes and favoritism, making the
process unfair. Checking resumes by hand takes a lot of time and can lead to errors,
and sometimes recruiters may unknowingly favor certain candidates. Many
existing tools only look for keywords, missing important skills.
This project introduces a Resume Analysis System using Machine Learning and
Natural Language Processing (NLP) in Python. It helps job seekers find the best
opportunities by matching their skills with suitable job roles. The system also gives
suggestions to improve resumes and allows users to link their profiles for better
recommendations. By using smart technology, it makes hiring more accurate, fair,
and efficient.
Existing Approaches/Related Works
Resume analysis and screening have been used for many years to select candidates for jobs. In the past
15 years, Machine Learning (ML) and Natural Language Processing (NLP) have improved this process
by making it faster, more accurate, and fairer.
• Nikolov et al. introduced methods to match candidates based on actual skills rather than false
experiences.
• Dutt improved resume screening with advanced ML algorithms, reducing bias and hiring time.
• Sinha developed the "Likelihoods Model" to highlight real skills and create more inclusive hiring.
• Brownlee enhanced TF-IDF, making job-role matching more accurate.
• Gupta & Raj improved resume analysis success by 19.5%, increasing efficiency.
• Manning et al. used NLP and ML to match resumes with job descriptions.
• Roy & Kinge proposed ML models to reduce hiring costs and speed up job matching.
Problems in Existing Approaches
•Resumes come in various formats (PDF, Word, plain text), making extraction difficult.
•Traditional systems struggle to identify key details like skills, experience, and education due to
unstructured data.
•Missing or incorrect information affects job matching accuracy.
•Many existing systems lack accuracy in analyzing resumes.
•Misinterpretation of job titles, skills, and experience levels leads to incorrect matches.
•Qualified candidates may be overlooked, while irrelevant resumes get selected.
•Overall, poor accuracy reduces hiring efficiency and fairness in candidate selection.
Proposed Methodology
The system uses Machine Learning (ML) and Natural Language Processing (NLP) to
improve resume screening and job matching.
1. Resume Preprocessing – Cleans resumes by removing unwanted data (links, special
characters, etc.).
2. Feature Extraction – Uses TF-IDF and NLP to extract key details like skills, experience,
and education.
3. Job Matching – ML algorithms compare resumes with job descriptions to find the best
fit.
4. Personalized Recommendations – Suggests ways to improve resumes and matches
candidates to better job roles.
5. Fair & Efficient Hiring – Reduces bias, speeds up recruitment, and ensures skill-based
selection.
Results and Discussion
•The resume dataset was divided into training and testing subsets to improve prediction accuracy.
•TF-IDF was used to convert resumes into numerical data for better analysis.
•Different machine learning models were tested for resume screening.
•K-Nearest Neighbor (KNN) achieved the highest accuracy of 98.44%, making it the best model.
•Other models, including One vs Rest and Support Vector Machine (SVM), also showed accuracy
improvements.
•The system ensures better job matching, reduced bias, and improved efficiency in the hiring process.
Conclusions and Future Work
•This research presents a resume analysis system using machine learning to predict job roles.
•The system accurately parses and evaluates resumes using NLP, making it efficient for large-scale
hiring.
•Automation in hiring reduces manual effort and speeds up the recruitment process.
•The system is designed to handle large volumes of job applications efficiently.
•Future improvements include supporting more file formats (Word, Google Docs, plain text) for better
accessibility.
•Enhancements will focus on making the platform more secure, user-friendly, and widely accessible.
References
1. Aleem, A., Kumar, A., & Gore, M.M. (2019). A Study of Manuscripts Evolution for Perfection –
ICACSE2019.
2. Nikolov, T., Chen, K., Corrado, G., & Dean, J. (2013). Efficient Estimation of Word Representations in
Vector Space – arXiv:1301.3781.
3. Dutt, S., Chandramouli, S., & Das, K. A. (2006). Machine Learning – Pearson.
4. Sinha, A. K., & M. A. Resume Screening Classification – DOI Reference.
5. Brownlee, J. (2016). Master Machine Learning Algorithms – Machine Learning Mastery.
6. Gupta, N. K., & Raj, A. Automatic Resume Filtering Using Machine Learning.
7. Manning, C. D., Raghavan, P., & Schütze, H. (2008). Introduction to Information Retrieval – Cambridge
University Press.
8. Aleem, A., Tiwary, P., Karn, S., & Kumar, V. (2022). Entertainment Advisor Using Sentiment Analysis –
Journal of Scientific Research, 66(2), 101-107.
9. Kinge, B., & S. M. Resume Screening Using Machine Learning (Vol. 253).
10. Roy, P. K., & S. S. Machine Learning Approach for Automation of Resume Recommendation System
(Vol. 167).
