Title: Exploring Uber’s Customer Feedback

Subtitle: Sentiment Analysis of 12,000+ Google Play Reviews: Unveiling Trends and Insights

Dataset Description

The dataset contains detailed review information, including unique reviewer usernames (over 12,000 entries), their profile pictures, and the content of their feedback, categorized as "good" (8%), "nice" (3%), and "other" (89%). Reviews are rated on a scale, with "ThumbsUp" indicating the count of likes (e.g., 239). Reviews are timestamped with creation dates, primarily "4.554.10001" (27%) and "4.555.10003" (19%), while other versions make up 54%. Each review includes the number of replies and the timestamp of the latest reply. Additionally, the app version being reviewed follows a similar distribution pattern, with key versions dominating a minority of the data.

Applications of the Uber Customer Reviews Dataset:

• Sentiment Analysis: Analyzing customer emotions (positive, negative, neutral) about the app.
• Natural Language Processing (NLP): Developing models to extract insights and key themes from text reviews.
• Service Improvement: Identifying common issues and user suggestions for app enhancement.
• Market Analysis: Studying trends in user feedback to understand market expectations.
• Machine Learning Models: Training algorithms to classify or predict user satisfaction based on reviews.

Column Indicators

• UserName: Name or alias of the user who provided the review.
• UserImage: Profile picture or avatar of the reviewer.
• Content: Text of the user's review or feedback.
• Score: Rating given by the user (e.g., out of 5 stars).
• ThumbsUp: Count of likes or approvals received for the review.
• ReviewCre: Date and time when the review was created.
• At: Timestamp of when the review was recorded or published.
• ReplyCount: Number of replies or comments on the review.
• RepliedAt: Timestamp of the most recent reply to the review.
• AppVersion: Version of the application being reviewed.