Sentiment Analysis for Staff Recruitment

Abstract

Purpose: This project aims to enhance recruitment processes by applying sentiment analysis to job descriptions from Naukri.com, categorizing them as positive, neutral, or negative. This method provides HR professionals with deeper insights into the sentiment conveyed in job postings. Design/Methodology/Approach: The methodology involves several steps: data preprocessing, including text conversion to lowercase, noise removal, tokenization, lemmatization, and stop word elimination; sentiment analysis using TextBlob to compute polarity scores; and categorization of these scores into sentiment classes. TF-IDF vectorization is used to convert the text input into numerical features, which are then used to train an SVM (Support Vector Machine) classifier. Metrics including accuracy, precision, recall, and F1-score are used to evaluate the performance of the model. Findings: The project demonstrates that sentiment analysis can significantly refine the recruitment process by filtering out less relevant job descriptions and highlighting those that match the company's needs and values. This approach aids in identifying job postings that are more likely to attract highly skilled candidates efficiently. Originality/Value: This study emphasizes the innovative use of sentiment analysis in hiring, with a particular emphasis on job descriptions. Through the application of NLP techniques and machine learning models, the project provides an approachable framework that recruiting processes may be customized and adopted by enterprises