metadata

annotations_creators: []
language_creators: []
language:
  - si
  - en
license:
  - mit
multilinguality:
  - multilingual
size_categories: []
source_datasets: []
task_categories:
  - text-classification
task_ids:
  - sentiment-analysis
  - hate-speech-detection
  - humor-detection
  - language-identification
  - aspect-identification

Sinhala-English-Code-Mixed-Code-Switched-Dataset

This dataset contains 10,000 comments that have been annotated at the sentence level for sentiment analysis, humor detection, hate speech detection, aspect identification, and language identification.

The following is the tag scheme.

Sentiment - Positive, Negative, Neutral, Conflict
Humor - Humorous, Non humorous
Hate Speech - Hate-Inducing, Abusive, Not offensive
Aspect - Network, Billing or Price, Package, Customer Service, Data, Service or product, None
Language ID - Sinhala, English, Sin-Eng, Eng-Sin, Mixed, Named-Entity, Symbol