|
|
--- |
|
|
license: apache-2.0 |
|
|
--- |
|
|
Finetuned BERT for Tech Product Names Named Entity Recognition (NER) |
|
|
|
|
|
GitHub: https://github.com/ashleyliu31/finetuned_bert_for_ner |
|
|
|
|
|
This NER model can recognize and tag tech product names like 'Asus ZenBook UX430UN', 'Acer Aspire 3', 'Nokia 110 4G', or 'Xiaomi 11T Pro 5G Hyperphone' in a sentence. The model was trained on the names of laptops and mobile phones. It might not be suitable for other tech products. |
|
|
|
|
|
To test the model, enter a sentence that contains a laptop or mobile phone product name in the "Hosted inference API" input field and press "Compute". The model will highlight and tag the product name in the sentence. |
|
|
|
|
|
Sample sentence to enter: "I love my new Razer Blade 16." "How much is the new IPhone 16 Pro Max?" |
|
|
|
|
|
Evaluation |
|
|
|
|
|
Evaluation code: https://colab.research.google.com/drive/19lR8KIxQ5DzSuJNI3b1sIMmfS48Br4Ln?usp=sharing |
|
|
|
|
|
Classification Report: |
|
|
|
|
|
precision recall f1-score support |
|
|
|
|
|
B-pn 0.96 0.95 0.96 110 |
|
|
I-pn 0.96 0.97 0.96 264 |
|
|
O 0.99 0.99 0.99 876 |
|
|
|
|
|
|
|
|
accuracy 0.98 1250 |
|
|
macro avg 0.97 0.97 0.97 1250 |
|
|
weighted avg 0.98 0.98 0.98 1250 |
|
|
|
|
|
|
|
|
Test data: https://huggingface.co/datasets/ashleyliu31/bert-ner-test-data |
