Smart Resume Analyzer: An NLP-Based System for Automated Resume Parsing and Job-Fit Scoring

This paper presented the design and implementation of the Smart Resume Analyzer, an NLP-based system that automates resume parsing, job-fit scoring, and candidate feedback generation. The proposed system integrates a fine-tuned spaCy NER pipeline, a multi-layer skills normalization taxonomy, and a weighted TF-IDF cosine similarity scoring algorithm within a scalable three-tier web architecture comprising React, FastAPI, and a Celery-based asynchronous processing engine. A key contribution of the proposed system is its ability to combine accurate, explainable job-fit scoring with structured, prioritized feedback — a capability largely absent from both academic resume-parsing research and commercial ATS tools. Evaluation results demonstrate that the system achieves strong correlation with human recruiter judgment (r = 0.83) while maintaining response times suitable for real-time interactive use, and that the generated feedback is independently rated as useful by the large majority of candidate users surveyed. By providing transparent, data-driven insight into resume-job alignment, the system has the potential to improve both recruitment efficiency and candidate experience. Future work will explore Sentence-BERT-based semantic scoring to close the remaining gap with human evaluators, multilingual resume support for Hindi and Marathi, depth-of-experience modeling that accounts for the duration associated with each skill, and direct integration with enterprise ATS platforms such as Greenhouse and Lever.

[1]  J. Lafferty, A. McCallum, and F. C. N. Pereira, “Conditional random fields: Probabilistic models for segmenting and labeling sequence data,” in Proc. 18th Int. Conf. Machine Learning (ICML 2001), Williamstown, MA, USA, Jun. 2001, pp. 282–289. [2]  T. Yu, R. Li, Z. Liu, A. Qi, J. Ye, and Z. Liu, “Resume information extraction with cascaded hybrid model,” in Proc. 43rd Annu. Meeting Assoc. Computational Linguistics (ACL 2005), Ann Arbor, MI, USA, Jun. 2005, pp. 499–506. [3]  G. Lample, M. Ballesteros, S. Subramanian, K. Kawakami, and C. Dyer, “Neural architectures for named entity recognition,” in Proc. 2016 Conf. North American Chapter Assoc. Computational Linguistics (NAACL-HLT 2016), San Diego, CA, USA, Jun. 2016, pp. 260–270. [4]  J. Devlin, M. W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” in Proc. 2019 Conf. North American Chapter Assoc. Computational Linguistics (NAACL-HLT 2019), Minneapolis, MN, USA, Jun. 2019, pp. 4171–4186. [5]  D. Shen, R. Zhu, C. Zhu, W. Qin, C. Wang, and H. Xiong, “A joint learning approach to intelligent job interview assessment,” in Proc. 27th Int. Joint Conf. Artificial Intelligence (IJCAI-18), Stockholm, Sweden, Jul. 2018, pp. 3542–3548. [6]  N. Reimers and I. Gurevych, “Sentence-BERT: Sentence embeddings using siamese BERT-networks,” in Proc. 2019 Conf. Empirical Methods Natural Language Processing (EMNLP-IJCNLP 2019), Hong Kong, China, Nov. 2019, pp. 3982–3992. [7]  S. Robertson and H. Zaragoza, “The probabilistic relevance framework: BM25 and beyond,” Foundations and Trends in Information Retrieval, vol. 3, no. 4, pp. 333–389, 2009. [8]  M. Honnibal and I. Montani, “spaCy 2: Natural language understanding with Bloom embeddings, convolutional neural networks and incremental parsing,” 2017. [Online]. Available: https://spacy.io [9]  S. Rose, D. Engel, N. Cramer, and W. Cowley, “Automatic keyword extraction from individual documents,” in Text Mining: Applications and Theory, M. W. Berry and J. Kogan, Eds. Chichester, UK: John Wiley & Sons, 2010, ch. 1, pp. 1–20. [10] Society for Human Resource Management (SHRM), “SHRM Talent Acquisition Benchmarking Report 2023,” SHRM, Alexandria, VA, USA, Tech. Rep., 2023.