1 of 11

Word-level Language Identification in Code-mixed Dravidian Languages

Harshitha S Kumar, Sharal Coelho, Asha Hegde, Kavya G and

Hosahalli Lakshmaiah Shashirekha

Department of Computer Science, Mangalore University

16th Forum for Information Retrieval Evaluation (FIRE 2024)

​

2 of 11

Introduction

Word-level Language Identification (LI)

It aims to identify the language of individual words within a given sentence.

​

In this paper, we - team MUCSNLPLab, describe the Conditional Random Field (CRF) model trained with text-based features to identify the language of the word in a given code-mixed text.

​

Word-level LI task is modeled as a sequence labeling problem where the goal is to assign a label to each word in a sequence.

​

​

​

3 of 11

Task Description

  • To develop models for word-level LI using the CoLI-Dravidian datasets.
  • This dataset consists of code-mixed YouTube comments in four Dravidian languages: Tamil, Kannada, Malayalam, and Tulu.

​

4 of 11

5 of 11

Methodology

​

This is achieved by training CRF model with text based features like next words, previous words, etc.

​

The steps involved in the framework are described in the following :

​

  • Data Preparation and Feature Extraction: The data preparation phase necessitates a careful examination of the dataset and arranging the data to fit the selected feature extraction technique to extract text features that can be used to train the CRF model.

​

  • Model Construction: In this work, CRFSuite - a wrapper built using sklearn_crfsuite library, providing a scikit-learn compatible estimator for CRF is employed for implementing CRF.

​

​

6 of 11

Figure 1: Framework of CRF model

7 of 11

To capture the dependencies and relationships between the words in

the given sequence

8 of 11

9 of 11

10 of 11

Conclusion & Future Work

  • In this paper, we describe a CRF model submitted to ”CoLI-Dravidian 2024: Word-level Code-Mixed Language Identification in Dravidian Languages” shared task at FIRE 2024, for word-level LI in four code-mixed Dravidian languages (Kannada, Tulu, Tamil, and Malayalam).

​

  • By training the CRF model with text-based features, the proposed model obtained Macro F1 score of 0.608 for Tamil and secured 3rd rank.

​

  • Efficient techniques will be explored in the future to handle the imbalanced dataset and improve the performance of the proposed models

11 of 11

Thank You