Legal Entity Extraction: An Experimental Study of NER Approach for Legal Documents

In legal domain Name Entity Recognition serves as the basis for subsequent stages of artificial intelligence. this paper, authors have developed a dataset training (NER) in Indian domain. As first step research methodology study is done to identify and establish more entities than commonly used named such person, organization, location, so on. The annotators can make use these annotate different types documents. Variety text annotation tools are existence finding best one difficult task, experimented with various before settling on work. resulting annotations from unstructured be stored into JavaScript Object Notation (JSON) format which improves data readability manipulation simple. After annotation, contains approximately 30 documents 5000 sentences. This further train spacy pre-trained pipeline predict accurate name entities. accuracy names increased if models fine-tuned using texts.