Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM65630
Volume:
10 |
6 |
Jun 2026
Pages:
1-3
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Department of Computer Application
Nehru Institute of Information Technology And Management
Department of Computer Application
Nehru Institute of Information Technology And Management
Traditional job portals rely on keyword matching and manual filtering, leaving significant gaps between candidate skills and employer requirements. This paper presents AIRTJMP, a modular Python-based platform integrating resume parsing, machine-learning skill-gap analysis, semantic job-candidate matching, and real-time recommendation delivery. Built on six candidate attributes (skills, experience, education, location, salary expectation, availability), a BERT fine-tuned semantic encoder achieves 91.8% top-5 match precision (NDCG@5 = 0.89). A collaborative-filtering module improves cold-start recommendations, and a Kafka-driven pipeline delivers match updates within 2.1 seconds of new postings. A React dashboard renders ranked job cards, skill-gap reports, and employer analytics. Real-time matching reduces candidate time-to-apply by 34.2% versus keyword-only portals.
KEYWORDS
AIRTJMP integrates structured recruitment analytics, BERT semantic matching (91.8% P@5, NDCG@5 = 0.89), collaborative-filtering cold-start handling, and Kafka real-time streaming (≤2.1 s) into one production-ready platform, reducing candidate time-to-apply by 34.2% over keyword-only systems. It contributes a documented analytics framework, a realistic structured dataset, and an educational codebase spanning NLP, recommendation systems, and real-time stream processing.
[8] Johnson et al., FAISS: Efficient similarity search, arXiv:1702.08734,2021. [9] Chawla et al., JAIR 16, 321-357, 2002. [10] Pedregosa et al., JMLR 12, 2825-2830, 2011. Acknowledgment: The author thanks the MCA faculty of Nehru Institute of Information Technology and Management, Coimbatore, and mentor Mr. Sivakumar P, along with the HuggingFace, FAISS, Apache Kafka, scikit-learn, and Flask open-source communities.
Publication Date
6/28/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
6
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...