Resume Classification Using LGBM Algorithm with Sentiment Analysis

The paragraph discusses the challenge of automating the classification of resumes, emphasizing the limitations of traditional manual methods due to the unstructured nature and diverse content found in resumes. To overcome these challenges, the article proposes the use of machine learning techniques for resume classification. Machine learning models can be trained on a large dataset of categorized resumes to identify patterns associated with different types of resumes, ultimately enabling faster, more precise, and impartial classification compared to manual methods. The proposed methodology combines sentiment analysis with the light gradient boosting machine (LGBM) algorithm to achieve this goal. The process involves several key steps, beginning with the collection of a dataset of resumes that have already been categorized into relevant job titles or categories. Next, the text within these resumes is preprocessed, which includes removing stop words and other irrelevant noise. Features are then extracted from the text, including its tone, keyword frequency, and length. The LGBM model is subsequently trained using this preprocessed resume dataset. The results of this approach are promising. The LGBM model exhibits an impressive 84% accuracy in categorizing resumes based on job titles, while sentiment analysis attains a perfect accuracy rate of 100%. Combining these two techniques results in a 1% increase in overall accuracy, demonstrating the effectiveness of this combined approach for resume classification.