Doctoral Student Vrunda Gadesha Presents New Approach to Automatically Annotate Text Data Sets

Vrunda Gadesha, a doctoral student at the School of Engineering and Applied Sciences, provides a skilful approach for automatically annotating text data sets, reducing reliance on manual sorting and labelling in her paper 'Streamlining Resume Annotation with Named Entity Recognition.''

She presented her paper at the Indian Council International Conference (INDICON) 2023. Her paper introduces a novel method for efficient resume annotation in IT, higher education, law, and finance fields. It utilises Named Entity Recognition (NER) to address the challenges of manual data processing and propose an automated, accurate solution pivotal for natural language processing applications.

In many industries, the resumes flow in high volume but are cluttered. Creating large-scale text data sets of resumes has traditionally relied on manual sorting and labelling, making it time-consuming and prone to inconsistencies due to human intervention. Consequently, these data sets may not fulfil the requirements of the actual industrial applications. Vrunda's research addresses these challenges with an approach for automatically annotating text data sets, which includes the combination of bi-LSTM and CRF, a powerful assistant in labelling the data with multiple entities.