Customer Complaints Sentiment And Priority Dataset

Context

These are real-world complaints regarding financial products and services. Each complaint has been labeled with its respective product category, sentiment, and priority level to facilitate the analysis of customer feedback. This dataset offers a structured methodology for comprehending customer issues and prioritizing responses. It is especially valuable for machine learning practitioners and researchers who specialize in sentiment analysis, natural language processing, and optimizing customer service.

Content

This dataset is an extension of an original financial dataset sourced from Kaggle. Here are the key details:

• Number of Records: 1750
• Classes: 7 product classes with 250 records each
• Attributes: Each record includes:
  • Product: The financial product category
  • Consumer Complaint: The text of the customer complaint
  • Sentiment: Manually labeled sentiment (0 for neutral, -1 for negative)
  • Priority: Manually labeled priority (0 for not urgent, 1 for urgent)

Modifications and Additions:

• Selection: Chosen 250 records per product class
• Manual Labeling: Sentiment and priority values were manually labeled based on the following criteria:
  • Sentiment Values: 0 for neutral, -1 for negative (no positive values due to the nature of complaints)
  • Priority Values: 0 for not urgent, 1 for urgent

Acknowledgements

We extend our gratitude to the creators of the original financial dataset on Kaggle. Their comprehensive data provided a solid foundation for this project. The original dataset can be accessed here: Consumer Complaint Database.

This extended dataset is licensed under the CC BY 4.0 International License. You are free to use, share, and adapt the data as long as you give appropriate credit, provide a link to the license, and indicate if changes were made.