Summary of L3cube-mahasocialner: a Social Media Based Marathi Ner Dataset and Bert Models, by Harsh Chaudhari et al.
L3Cube-MahaSocialNER: A Social Media based Marathi NER Dataset and BERT models
by Harsh Chaudhari, Anuja Patil, Dhanashree Lavekar, Pranav Khairnar, Raviraj Joshi
First submitted to arxiv on: 30 Dec 2023
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper presents the L3Cube-MahaSocialNER dataset, a large-scale social media dataset designed specifically for Named Entity Recognition (NER) in Marathi language. The dataset consists of 18,000 manually labeled sentences covering eight entity classes, addressing challenges posed by social media data such as non-standard language and informal idioms. The authors evaluate deep learning models like CNN, LSTM, BiLSTM, and Transformer on this dataset with IOB and non-IOB notations. Results show that these models accurately recognize named entities in Marathi informal text. The L3Cube-MahaSocialNER dataset provides user-centric information extraction and supports real-time applications, making it a valuable resource for public opinion analysis, news, and marketing on social media platforms. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research creates a big social media data set to help machines recognize important words and phrases in Marathi language. Marathi is spoken by people in India. The data set has 18,000 sentences that are labeled as correct or not. It’s hard for machines to understand informal language used on social media, so this data set helps with that problem. The researchers tested different machine learning models on the data and found that they work well. This means that machines can now better understand important information from Marathi social media posts. |
Keywords
* Artificial intelligence * Cnn * Deep learning * Lstm * Machine learning * Named entity recognition * Ner * Transformer
