Few-shot Named Entity Recognition: Definition, Taxonomy and Research Directions

Abstract

Recent years have seen an exponential growth (+98% in 2022 w.r.t. the previous year) of the number of research papers in the few-shot learning field, which aims to train machine learning models with extremely limited available data. The research interest towards few-shot learning systems for Named Entity Recognition (NER) is thus at the same time increasing. NER consists in identifying mentions of pre-defined entities from unstructured text, and serves as a fundamental step in many downstream tasks, such as the construction of Knowledge Graphs, or Question Answering. The need for a NER system able to be trained with few-annotated examples comes in all its urgency in domains where the annotation process requires time, knowledge and expertise (e.g. healthcare, finance, legal), and in low-resource languages. In this survey, starting from a clear definition and description of the few-shot NER (FS-NER) problem, we take stock of the current state-of-the-art and propose a taxonomy which divides algorithms in two macro-categories according to the underlying mechanisms: model-centric and data-centric. For each category, we line-up works as a story to show how the field is moving towards new research directions. Eventually, techniques, limitations and key aspects are deeply analyzed to facilitate future studies.

State-of-art approach for Indian Language based on NER: Comprehensive Review

“…(i) Low Resources Languages: Some Indian Languages have very limited digital content and resources, making it challenging to train and deploy NER models [45]. When rules are excessively detailed, it may result in inaccurate categorization.…”

“…(b) Question Answering Systems: Named Entity Recognition helps in identifying entities mentioned in a question and linking them to relevant information in a knowledge base. This enables question answering systems to provide more accurate and relevant answers [4], [12], [39], [41], [42], [44], [45]. (e) Social Media Analysis: NER is valuable for analyzing social media data.…”

Semi-automatic Knowledge Graph Construction Based on Deep Learning

“…Nowadays, information technology in education has received unprecedented attention and rapid development, which has brought new technical support and development opportunities for curriculum teaching reform in the new environment [1][2][3].…”