Machine Learning based Custom Named Entity Recognition
Akula V. S. Siva Rama Rao, Akula Pavan Sreevatsav · 2022 International Conference on Augmented Intelligence and Sustainable Systems (ICAISS) · 2022
The significant increase of digital data in real world also results in various challenges in extracting the intended entities from the input text. Entities extraction process plays an important role in Natural Language Processing (NLP) applications such as named entity recognition, machine translation, text analytics, parts of speech tagging, sentiment analysis etc. In the real world, there are huge number of Named Entities, the recent named entity recognizer, SpaCy API version 3.0 can extract eighteen types of entities. In order to create and extract the user defined custom entities, this research work has proposed a Custom Named Entity Recognition (CNER) model. This model uses few textual training datasets by including user defined custom entities. By using CNER model, ten user intended custom named entities were trained and extracted from 27,439 English words present in eight different domains and Wikipedia articles and finally the state-of-the-art performance results are obtained.