Development of a Dataset and a Deep Learning Baseline Named Entity Recognizer for Three Low Resource Languages: Bhojpuri, Maithili and Magahi

TL;DR

This paper creates a NER dataset for Bhojpuri, Maithili, and Magahi, low-resource languages, and establishes a deep learning baseline using LSTM-CNNs-CRF models to improve entity recognition accuracy.

Contribution

It introduces the first annotated NER benchmark dataset for three low-resource languages and evaluates a deep learning baseline for NER in these languages.

Findings

Deep learning baseline achieves comparable or better F1-scores than CRF models.

Annotated corpora with 22 entity labels for each language.

Provides a foundation for future NLP tasks in low-resource languages.

Abstract

In Natural Language Processing (NLP) pipelines, Named Entity Recognition (NER) is one of the preliminary problems, which marks proper nouns and other named entities such as Location, Person, Organization, Disease etc. Such entities, without a NER module, adversely affect the performance of a machine translation system. NER helps in overcoming this problem by recognising and handling such entities separately, although it can be useful in Information Extraction systems also. Bhojpuri, Maithili and Magahi are low resource languages, usually known as Purvanchal languages. This paper focuses on the development of a NER benchmark dataset for the Machine Translation systems developed to translate from these languages to Hindi by annotating parts of their available corpora. Bhojpuri, Maithili and Magahi corpora of sizes 228373, 157468 and 56190 tokens, respectively, were annotated using 22 entity labels. The annotation considers coarse-grained annotation labels followed by the tagset used in one of the Hindi NER datasets. We also report a Deep Learning based baseline that uses an LSTM-CNNs-CRF model. The lower baseline F1-scores from the NER tool obtained by using Conditional Random Fields models are 96.73 for Bhojpuri, 93.33 for Maithili and 95.04 for Magahi. The Deep Learning-based technique (LSTM-CNNs-CRF) achieved 96.25 for Bhojpuri, 93.33 for Maithili and 95.44 for Magahi.