README.md

metadata

language:
  - en
license: apache-2.0
size_categories:
  - 10K<n<100K
task_categories:
  - conversational
  - text-generation
  - summarization
  - question-answering
  - text-classification
  - text-retrieval
  - translation
pretty_name: No Robots
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
dataset_info:
  features:
    - name: title
      dtype: string
    - name: content
      dtype: string
    - name: author
      dtype: string
    - name: category
      dtype: string
    - name: published_date
      dtype: string
    - name: page_url
      dtype: string
  splits:
    - name: train
      num_bytes: 81377758
      num_examples: 30374
    - name: test
      num_bytes: 12272769
      num_examples: 9489
  download_size: 0
  dataset_size: 93650527
tags:
  - ghana
  - news
  - politics
  - science
  - business
  - ghana-news

Description 🙅‍♂️🤖

GhanaNews dataset is a collection of news articles from various Ghanaian News Portals (MyJoyOnline, GraphicOnline, GhanaWeb, PulseGh, CitiNewsOnline, ect). The dataset is provided by the academic comunity for research purposes in data mining (clustering, classification, etc), information retrieval (ranking, search, etc), xml, data compression, data streaming, and any other non-commercial activity.

The Ghana news topic classification dataset is constructed by Theophilus Siameh ([email protected]) from the dataset above.

Context QA: in context question answering from an article

{"article": "...", "question": "...", "answer": "..."}

Article and Summary

{"article": "...", "summary": "..."}

Dataset Format

{
    "title": "...", 
    "content": "...", 
    "author": "...", 
    "category": "...", 
    "published_date": "...", 
    "page_url": "..."
}

Load Dataset

pip install datasets

from datasets import load_dataset

train = load_dataset("worldboss/ghana-news", split="train")
test =  load_dataset("worldboss/ghana-news", split="test")

pd.DataFrame(train).head()