Resumate: A Prototype to Enhance Recruitment Process with NLP based Resume Parsing

The process of reviewing resumes takes time. The unstructured written language can be understood and parsed by natural language processing and machine learning, which can then extract the required information. The goal is to teach the computer to analyze written papers similarly to a human. For assessing and analyzing the unstructured data, various methods including named entity recognition, tokenization, text classification, and approaches of other natural language processing techniques have been explored. A user uploads a resume into the program, which takes it and converts it into a standard text format before parsing it for the necessary information and organizing the extracted data in a standard defined format. After extraction and preprocessing, a predictive model is trained is better and categorize resumes for use in various applications. The k-fold cross validation with k=5 outcome in this case reveals that SVM(0.58 - 0.65) and XG Boost(0.65 - 0.73) have higher ranges of training performance than all other models. Here, the XG Boost outperforms previous trained models, accurately classifying the newly created resume into the target category. The proposed work is novel and presents one of its kind methodology.