ismailduru/Telco-DPR

---
license: apache-2.0
configs:
- config_name: corpus
data_files:
- split: small
path: corpus/corpus-00000-of-00001.parquet
- split: extended
path: corpus/extended-00000-of-00001.parquet
- config_name: queries
data_files:
- split: queries
path: queries/queries-00000-of-00001.parquet
- config_name: relevant_docs
data_files:
- split: test
path: relevant_docs/test-00000-of-00001.parquet
- split: train
path: relevant_docs/train-00000-of-00001.parquet

---
# Dataset Information
This paper proposes a Question-Answering (QA) system for the telecom domain using 3rd Generation Partnership Project (3GPP) technical documents.
Alongside, a hybrid dataset, Telco-DPR, which consists of a curated 3GPP corpus in a hybrid format, combining text and tables, is presented.
Additionally, the dataset includes a set of synthetic question/answer pairs designed to evaluate the retrieval performance of QA systems on this type of data.
The retrieval models, including the sparse model, Best Matching 25 (BM25), as well as dense models, such as Dense Passage Retriever (DPR)
and Dense Hierarchical Retrieval (DHR), are evaluated and compared using top-K accuracy and Mean Reciprocal Rank (MRR).
The results show that DHR, a retriever model utilising hierarchical passage selection through fine-tuning at both the document and passage levels,
outperforms traditional methods in retrieving relevant technical information, achieving a Top-10 accuracy of 86.2%.
Additionally, the Retriever-Augmented Generation (RAG) technique, used in the proposed QA system,
is evaluated to demonstrate the benefits of using the hybrid dataset and the DHR. The proposed QA system, using the developed
RAG model and the Generative Pretrained Transformer (GPT)-4, achieves a 14% improvement in answer accuracy, when compared to a previous benchmark on the same dataset.
https://arxiv.org/abs/2410.19790

### Python Code to Load Dataset

### Git Clone Load Dataset
-> git clone https://huggingface.co/datasets/thainasaraiva/Telco-DPR

```python
from datasets import concatenate_datasets, load_dataset, DatasetDict

corpus_ds = load_dataset('parquet', data_dir='./Telco-DPR/corpus', data_files={"corpus":'corpus-00000-of-00001.parquet'})
corpus_extend_ds = load_dataset('parquet', data_dir='./Telco-DPR/corpus', data_files={"extended":'extended-00000-of-00001.parquet'})
corpus_ds=DatasetDict({
'corpus': concatenate_datasets([corpus_ds['corpus'],corpus_extend_ds['extended']])
})

queries_ds = load_dataset('parquet', data_dir='./Telco-DPR/queries', data_files={"queries":'queries-00000-of-00001.parquet'})

relevant_docs_ds = load_dataset('parquet', data_dir='./Telco-DPR/relevant_docs', data_files={"train":'train-00000-of-00001.parquet',"test":'test-00000-of-00001.parquet'})