krusty99/sentiment-analysis-dataset

license: mit task_categories:

• —text-classification language:
• —en tags:
• —finance prettyname: sentiment-analysis-dataset sizecategories:
• —n<1K ---

Dataset Card for Sentiment Analysis Dataset

This dataset card aims to provide a comprehensive overview of a sentiment analysis dataset containing product reviews labeled with sentiment.

Dataset Details

Dataset Description

This dataset contains 1,000 product reviews categorized into two sentiment classes: Positive and Negative. It can be utilized for training and evaluating sentiment analysis models in natural language processing tasks.

• —Curated by: krusty99
• —Language(s) (NLP): English
• —License: MIT

Uses

Direct Use

This dataset is intended for users interested in building, training, or evaluating models for sentiment analysis in English text, particularly in the domain of product reviews.

Out-of-Scope Use

This dataset should not be used for tasks requiring high accuracy in contexts where sensitive decision-making is involved, such as healthcare or legal outcomes.

Dataset Structure

The dataset contains two fields:

• —review: A string containing the text of the product review.
• —sentiment: A categorical label indicating whether the sentiment is “Positive” or “Negative.”

Dataset Creation

Curation Rationale

The dataset was created to address the lack of labeled sentiment analysis datasets specific to product reviews, particularly those in the English language, facilitating the evaluation of sentiment prediction algorithms.

Source Data

The data was collected from publicly available product review websites, ensuring the datasets represent a variety of products and opinions.

Data Collection and Processing

Data were scraped using web scraping tools (e.g., Beautiful Soup, Scrapy) and processed to remove duplicates and irrelevant entries. Reviews were then labeled based on sentiment through manual annotation.

Who are the Source Data Producers?

The source data was originally produced by consumers who reviewed the products on various commercial websites.

Annotation Process

The annotation was performed by a team of three annotators. Each annotator received guidelines detailing how to classify reviews based on sentiment. The annotation process lasted two weeks, during which 1,000 reviews were labeled. Interannotator agreement was measured and found to be 85%, indicating good reliability.

Who are the Annotators?

The annotators were graduate students from a natural language processing program, trained in sentiment analysis and familiar with product reviews. Their demographic information, while not collected, includes diverse backgrounds.

Personal and Sensitive Information

The dataset does not contain any personal or sensitive information as it consists solely of public product reviews. All identifying information has been stripped away, and efforts to anonymize the data were followed.

Bias, Risks, and Limitations

While the dataset provides a foundation for sentiment analysis, it may be subject to biases based on the type of products reviewed and the demographics of the reviewers. Certain sentiments may be over- or under-represented based on review length, context, or specific product categories.

Recommendations

Users should be aware of potential bias introduced by the selected products and reviewers in the dataset. It is recommended to use this dataset in conjunction with other datasets to create a more balanced sentiment analysis model.

Citation

If you use this dataset, please cite it as follows:

BibTeX:

@dataset{sentiment_analysis,  
  title={Sentiment Analysis Dataset},  
  author={Krusty99},  
  year={2023},  
  publisher={Hugging Face},  
  url={https://huggingface.co/krusty99/sentiment-analysis-dataset}  
}