[Corpora-List] Call for participation-Sentiment Analysis of Dravidian Languages (Tamil, Malayalam and Kannada) in Code-Mixed Text 2021

Bharathi Raja Asoka Chakravarthi bharathi.raja at insight-centre.org
Thu Apr 22 09:26:13 CEST 2021


Call for participation

Track name

Sentiment Analysis of Dravidian Languages (Tamil, Malayalam, and Kannada) in Code-Mixed Text 2021

Codalab link: https://competitions.codalab.org/competitions/30642

Webpage link: https://dravidian-codemix.github.io/2021/index.html

Track description

Sentiment analysis is the task of identifying subjective opinions or emotional responses about a given topic. It has been an active area of research in the past two decades in both academia and industry. There is an increasing demand for sentiment analysis on social media texts which are largely code-mixed for Dravidian languages. Code-mixing is a prevalent phenomenon in a multilingual community and the code-mixed texts are sometimes written in non-native scripts. Systems trained on monolingual data fail on code-mixed data due to the complexity of code-switching at different linguistic levels in the text. This shared task presents a new gold standard corpus for sentiment analysis of code-mixed text in Dravidian languages (Tamil-English, Malayalam-English, and Kannada-English).

The Tamil language is spoken by Tamil people in India, Sri Lanka, and by the Tamil diaspora around the world, with official recognition in Tamil Nadu, India, Sri Lanka, and Singapore. Kannada and Malayalam are a Dravidian language spoken predominantly by the people of Karnataka, and Kerala, India. The Tamil script evolved from the Tamili script, Vatteluttu alphabet, and Chola-Pallava script. It has 12 vowels, 18 consonants, and 1 āytam (voiceless velar fricative). Minority languages such as Saurashtra, Badaga, Irula, and Paniya are also written in the Tamil script. Tamil scripts are explained in Tolkappiyam as Eluttu means "sound, letter, phoneme", and this covers the sounds of the Tamil language, how they are produced (phonology). It includes punarcci (lit. "joining, copulation") which is a combination of sounds, orthography, graphemic, and phonetics with sounds as they are produced and listened to. Both Kannada and Malayalam scripts are alpha-syllabic, belonging to a family of abugida writing systems that is partially alphabetic and partially syllable-based. However, social media users often mix Roman script for typing because it is easy to input. Hence, the majority of the data available in social media for these under-resourced languages are code-mixed.

The goal of this task is to identify the sentiment polarity of the code-mixed dataset of comments / posts in Tamil-English, Malayalam-English, and Kannada-English collected from social media. The comment / post may contain more than one sentence but the average sentence length of the corpora is 1. Each comment / post is annotated with sentiment polarity at the comment / post level. This dataset also has class imbalance problems depicting real-world scenarios. Our proposal aims to encourage research that will reveal how sentiment is expressed in code-mixed scenarios on social media.

The participants will be provided development, training, and test dataset.

Task A: This is a message-level polarity classification task. Given a Youtube comment, systems have to classify it into positive, negative, neutral, mixed emotions, or not in the intended languages.

Data

The data is in format as below

Comment label

Intha padam vantha piragu yellarum Thala ya kondaduvanga positive

Tamil-English: 44,020 comments, Train: 35,220 Validation: 4,398 and Test: 4,402

Malayalam-English: 19,616 comments, Train: 15,694 Validation: 1,960 and Test: 1,962

Kannada-English: 7,671 comments, Train: 6,136 Validation:767 and Test: 768

We present Tamil-English, Kannada-English, and Malayalam-English, a dataset of YouTube video comments. The dataset contains all three types of code-mixed sentences Inter-Sentential switch, Intra-Sentential switch, and Tag switching. Most comments were written in native script and Roman script with either Tamil / Malayalam / Kannada grammar with English lexicon or English grammar with Tamil / Malayalam / Kannada lexicon. Some comments were written in Tamil / Malayalam / Kannada script with English expressions in between.

Evaluation plan

The classification systems’ performance will be measured in terms of weighted averaged precision, weighted averaged recall, and weighted averaged F-Score across all the classes. Weighted averaged scores are averaged the support-weighted mean per label. Participants are encouraged to check their system with the Sklearn classification report

https://scikit-learn.org/stable/modules/generated/sklearn.metrics.classification_report.html

The participants are required to submit the predicted data in a tab separated single file named 'predictions.tsv’.

The ‘predictions.tsv’ file should have two columns named Comment (text), Sentiment Polarity (Positive, Negative, Neutral, Mixed feelings and Non-Tamil or Non-Malayalam or Non-Kannada)

Timeline:

- 15th April - open track websites and training data release

- 15th June – test data release

- 25th June – run submission deadline

- 15th July – results declared

- 31st August – Working notes due

- 15th Oct – Camera-ready copies of working notes and overview paper due

- Tentatively in 1st or 2nd week of December - FIRE 2021

Organizer/s Details:

-

Bharathi Raja Chakravarthi, Data Science Institute, National University

of Ireland Galway

-

Ruba Priyadharshini, ULTRA Arts and Science College, Madurai, Tamil Nadu

-

Sajeetha Thavareesan, Eastern University, Sri Lanka

-

Dhivya Chinnappa, Thomson Reuters, United States of America

-

John P. McCrae, Data Science Institute, National University of Ireland

Galway

-

Elizabeth Sherly, Indian Institute of Information Technology and

Management-Kerala, India

- Thenmozhi D, SSN College of Engineering, Tamil Nadu

Student Volunteer

-

Adeep Hande, Indian Institute of Information Technology Tiruchirappalli,

Tamil Nadu

-

Rahul Ponnsamy, Indian Institute of Information Technology and

Management-Kerala

-

Shubhanker Banerjee, National University of Ireland Galway

-

Vasantharajan Charangan, University of Moratuwa, Sri Lanka

Contact details:

Email: bharathiraja.akr at gmail.com, dravidiancodemixed at gmail.com

Codalab link: https://competitions.codalab.org/competitions/30642

Webpage link: https://dravidian-codemix.github.io/2021/index.html

with regards, Dr. Bharathi Raja Chakravarthi, Adjunct Lecturer at School of Computer Science, National University of Ireland Galway Postdoctoral Fellow at Unit for Linguistic Data, Insight SFI Research Centre for Data Analytics, Data Science Institute, National University of Ireland Galway https://github.com/dravidian-codemix/2021/ https://sites.google.com/view/lt-edi-2021/home https://dravidianlangtech.github.io/2021/ https://dravidian-codemix.github.io/2020/ E-mail: bharathiraja.akr at gmail.com <bharathi.raja at insight-centre.org> Web: http://www.nuigalway.ie:83/our-research/people/engineering-and-informatics/basokachakravarthi1/ Google Scholar: https://scholar.google.com/citations?user=irCl028AAAAJ&hl=en LinkedIn: https://www.linkedin.com/in/bharathi-raja-asoka-chakravarthi-7a520393/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 32382 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20210422/ab1c5ffe/attachment.txt>



More information about the Corpora mailing list