The viewer is disabled because this dataset repo requires arbitrary Python code execution. Please consider removing the loading script and relying on automated data support (you can use convert_to_parquet from the datasets library). If this is not possible, please open a discussion for direct help.

Indic Sentiment Analysis

Description

The Indic Sentiment Analysis dataset contains reviews from various categories and sub-categories in multiple Indic languages. Each review is labeled with sentiment polarity (positive, negative, or neutral).

Dataset Structure

Data Fields

  • CATEGORY: The broad category to which the review belongs.
  • SUB-CATEGORY: The sub-category within the main category.
  • PRODUCT: The specific product or service being reviewed.
  • BRAND: The brand associated with the product or service.
  • ASPECTS: Different aspects or features of the product or service being reviewed.
  • ASPECT COMBO: Combinations of aspects that are discussed in the review.
  • ENGLISH REVIEW: The review text in English.
  • LABEL: The sentiment label assigned to the review (positive, negative, or neutral).
  • INDIC REVIEW: The review text translated into various Indic languages.
Downloads last month
212

Models trained or fine-tuned on ai4bharat/IndicSentiment

Collections including ai4bharat/IndicSentiment