Skip to content

kytrungchauwork/ParallelSenseTagger

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 

Repository files navigation

Parallel Sense Tagger (RoBERTa-based)

Overview

Parallel Sense Tagger is a Natural Language Processing (NLP) project that performs Word Sense Disambiguation (WSD).
The system automatically assigns the correct semantic meaning to ambiguous words based on contextual information within a sentence.

This project focuses on improving efficiency by integrating parallel processing techniques during training and inference.


Model

The model is fine-tuned from roberta-base using Hugging Face Transformers and PyTorch.

🔗 Hugging Face Model:
https://huggingface.co/kytrungchauwork/eng-viet-parallel-sense-tagger

Pipeline tag: token-classification


Features

  • Fine-tuned RoBERTa model for token-level sense tagging
  • Context-aware word sense disambiguation
  • Parallelized data preprocessing and training
  • Scalable design for large text datasets

Tech Stack

  • Python
  • PyTorch
  • Hugging Face Transformers
  • Multiprocessing / Parallel Computing

Example Usage

from transformers import AutoTokenizer, AutoModelForTokenClassification

tokenizer = AutoTokenizer.from_pretrained("your-username/parallel-sense-tagger")
model = AutoModelForTokenClassification.from_pretrained("your-username/parallel-sense-tagger")

text = "The bank is near the river."
inputs = tokenizer(text, return_tensors="pt")
outputs = model(**inputs)

About

Developed a parallel word sense tagging system to disambiguate word meanings in context. Applied natural language processing techniques and parallel processing to improve tagging efficiency and performance.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors