|
--- |
|
license: mit |
|
--- |
|
|
|
# BanglaNER |
|
[](https://colab.research.google.com/drive/1uN1WP7MjaBYXKABfhkHGn7EBWm9kd9k9?usp=sharing) |
|
Bangla Name Entity Recognition (NER) is extracting human names from input Bangla string or text. To solve this problem select Spacy pipeline and try 5 experimental approaches. |
|
|
|
The experiment is done only using one entity name (person) labeled as PER. After completing the experiment we got the best performance from the spacy transformer-based model. |
|
|
|
For more detail please check the experimental details and Best model F1 score is ~.81.05. |
|
# Dataset |
|
|
|
Bangla NER data is collected from, |
|
|
|
1. banglakit Bangla NER Dataset [Link](https://raw.githubusercontent.com/banglakit/bengali-ner-data/master/main.jsonl) |
|
|
|
2. Rifat1493 Bangla NER Dataset [Link](https://github.com/Rifat1493/Bengali-NER/tree/master/Input) |
|
|
|
|
|
3. SemEval2022 Bangla NER Dataaset [Link](https://competitions.codalab.org/competitions/36425#learn_the_details) |
|
|
|
More Detail about the model check github. |
|
|
|
# Reference |
|
|
|
1. [Spacy Training Pipelines & Models](https://spacy.io/usage/training) |
|
2. [NER data annotation](https://doccano.github.io/doccano/tutorial/) |
|
|
|
3. [BERT Pretrin model ](https://github.com/csebuetnlp/banglabert) |
|
4. [BILOU data formats meaning](https://stackoverflow.com/questions/17116446/what-do-the-bilou-tags-mean-in-named-entity-recognition) |
|
5. [SpaCy 3.1 data format](https://zachlim98.github.io/me/2021-03/spacy3-ner-tutorial) |
|
6. [Tranformer infornation](https://jalammar.github.io/illustrated-transformer/) |
|
7. [Load Gensim WordVectors into spacy pipeline](https://stackoverflow.com/questions/75521069/load-gensim-wordvectors-into-spacy-pipeline) |
|
|
