Exploiting Bi-LSTMs for Named Entity Recognition in Indian Culinary Science

9 Pages Posted: 27 Feb 2020

See all articles by G.S. Mahalakshmi

G.S. Mahalakshmi

Anna University - Department of Computer Science and Engineering

Makesh Narsimhan Sreedhar

Anna University - Department of Computer Science and Engineering

Ravi Kiran Selvam

Anna University - Department of Computer Science and Engineering

S Sendhilkumar

Anna University - Department of Computer Science and Engineering

Date Written: February 27, 2020

Abstract

This paper discusses the use of Bidirectional LSTMs for recognition of Named Entities over the Indian Recipe Blogs. Recipe posts from popular blogs including Hebbar's Kitchen are harvested and trained for recognizing NEs. Both the word embeddings and character embeddings are utilized as feature vectors for training the Bi-LSTM. CRF model is used for joint decoding of the labels. The system shows a development data F1 score of 92.87% and test data F1 score of 94.66%. The dataset used and meta-results obtained are released freely for research use.

Keywords: Bi-LSTM, CNN, CRF, Named Entity Recognition, Trie of Words

Suggested Citation

Mahalakshmi, G.S. and Sreedhar, Makesh Narsimhan and Selvam, Ravi Kiran and Sendhilkumar, S, Exploiting Bi-LSTMs for Named Entity Recognition in Indian Culinary Science (February 27, 2020). 5th International Conference on Next Generation Computing Technologies (NGCT-2019), Available at SSRN: https://ssrn.com/abstract=3545088 or http://dx.doi.org/10.2139/ssrn.3545088

G.S. Mahalakshmi

Anna University - Department of Computer Science and Engineering ( email )

Chennai, Tamil Nadu
India

Makesh Narsimhan Sreedhar

Anna University - Department of Computer Science and Engineering ( email )

Chennai, Tamil Nadu
India

Ravi Kiran Selvam (Contact Author)

Anna University - Department of Computer Science and Engineering ( email )

Chennai, Tamil Nadu
India

S Sendhilkumar

Anna University - Department of Computer Science and Engineering ( email )

Chennai, Tamil Nadu
India

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
86
Abstract Views
437
rank
404,090
PlumX Metrics