Datasets by use case

Sentiment and Text Classification Datasets

Free text with a label to learn from. Reviews with star ratings and sentiment, social posts and comments with sentiment scores, news headlines by sector, chatbot messages with intent labels, and support tickets by category. Use them to fine-tune a classifier, test a prompt, or teach the basics of text analysis.

Reviews and social posts

Product reviews with a title, text and star rating (most with a sentiment label too), and social media posts with comments, each scored for sentiment. Compare what the star rating says with what the text says.

E-commerce17 cols · 200 rows

Customer Product Review Dataset

This dataset provides a comprehensive view of customer product reviews, including detailed ratings, review text, sentiment analysis, and user demographics such as age, gender, and location. It enables advanced sentiment analysis, natural language processing, and customer satisfaction research, making it ideal for businesses seeking to understand product perception and improve customer experience.

review_titleverified_purchasesentiment_label
Great Value!truepositive
Did not work as expectedtruenegative
It’s okayfalseneutral
review_idcustomer_idproduct_id+11 more
E-commerce12 cols · 200 rows

E-Commerce Reviews Sentiment Dataset

This dataset contains detailed e-commerce product reviews, including review text, star ratings, sentiment labels, and metadata such as helpful votes and verified purchase status. It is ideal for training and evaluating sentiment analysis models, understanding customer feedback, and powering recommendation engines. The rich structure supports both NLP research and practical business applications.

product_namereview_titlesentiment_label
Wireless Bluetooth EarbudsAmazing Sound Quality!positive
Stainless Steel Water BottleKeeps Drinks Cold For Hourspositive
Foldable Laptop StandMakes Working Easierpositive
review_idproduct_iduser_id+6 more
E-commerce13 cols · 200 rows

Product Reviews

This dataset provides detailed customer product reviews, including star ratings, review text, helpfulness votes, verified purchase status, and tracked responses from sellers or support teams. It enables comprehensive sentiment analysis, product feedback evaluation, and customer engagement tracking for e-commerce platforms.

review_titleverified_purchaseresponse_author
Absolutely fantastic!trueSupportTeam
Works as expectedtrue–
Not worth the pricetrueQA_Manager
review_idproduct_idcustomer_id+7 more
Social Media14 cols · 200 rows

Social Media Sentiment Dataset

This dataset contains synthetic social media posts and their associated comments, each labeled with sentiment (positive, negative, neutral) and optional sentiment scores. It is designed for training and benchmarking machine learning and NLP models in social analytics, enabling fine-grained sentiment analysis across multiple platforms and languages.

post_sentiment_labelcomment_sentiment_labelplatform
positivepositivetwitter
positivepositivefacebook
positivepositiveinstagram
post_iduser_idpost_datetime+8 more

Feedback, news and intents

Student course feedback, financial news headlines with sector and ticker, chatbot messages with intent labels and confidence scores, and user feedback labelled by theme and whether it can be acted on.

Education13 cols · 200 rows

Student Feedback Sentiment Analysis

This dataset provides detailed student feedback on courses, enriched with sentiment analysis labels and scores, course and instructor metadata, and student academic context. It enables educators and administrators to systematically analyze curriculum effectiveness, identify areas for improvement, and track sentiment trends over time for data-driven decision making.

sentiment_labelcourse_titleinstructor_name
positiveIntroduction to Computer ScienceDr. Alicia Thornton
neutralPrincipios de EconomíaProf. Ricardo Alvarez
neutral数据结构基础Dr. Li Wei
feedback_idstudent_idcourse_id+7 more
Finance13 cols · 200 rows

Financial News Sentiment Streams

This dataset aggregates real-time sentiment scores and metadata for financial news headlines, enabling rapid detection of market-moving events and trends. It includes headline text, publication details, sentiment analysis, relevance to financial markets, and links to affected stocks and sectors. Ideal for quantitative trading, risk monitoring, and financial news analytics.

source_namesentiment_labelevent_type
Reuterspositiveearnings
Handelsblattneutralmacroeconomic
CNBCpositiveearnings
headline_idheadline_textpublished_datetime+7 more
Healthcare11 cols · 200 rows

Healthcare Chatbot Intent Dataset

This dataset provides detailed, synthetic healthcare chatbot conversations with annotated intent labels, message sequencing, and extracted entities. Designed for training and evaluating conversational AI, it supports intent classification, dialogue modeling, and entity recognition in healthcare virtual assistants. The dataset enables robust analysis of user-bot interactions for improved patient engagement and automation.

sender_typeintent_labelconversation_topic
userbook_appointmentappointments
botrequest_timeappointments
userprovide_timeappointments
transcript_idmessage_idmessage_order+5 more
Technology12 cols · 200 rows

User Feedback Text Analysis Set

This dataset contains synthetic, multi-domain user feedback entries, each labeled for sentiment, theme, and actionability to support machine learning applications in automated response systems and product improvement. It enables granular analysis of user concerns, trends, and actionable insights across different product areas, making it ideal for customer experience optimization and workflow automation.

sentiment_labeltheme_labelactionability_label
positivefeature_requestactionable
negativeperformanceactionable
mixedsupportneeds_clarification
feedback_iduser_iddomain+6 more

Support tickets

IT tickets with a subject, description, category, priority and urgency: a realistic multi-class routing problem.

Technology20 cols · 200 rows

IT Service Ticket Classification

This dataset contains detailed records of IT service tickets, combining structured metadata (such as priority, category, and assignment) with rich ticket descriptions suitable for natural language processing. It enables automated ticket triage, prioritization, and advanced analytics for IT support operations, making it ideal for machine learning and process optimization.

ticket_subjectcategorypriority
Laptop battery fails to chargehardwaremedium
Cannot print from office printerhardwarehigh
Server room temperature alerthardwarecritical
departmentstatusurgency+14 more

Every dataset in this collection

9 datasets

E-commerce17 cols · 200 rows

Customer Product Review Dataset

This dataset provides a comprehensive view of customer product reviews, including detailed ratings, review text, sentiment analysis, and user demographics such as age, gender, and location. It enables advanced sentiment analysis, natural language processing, and customer satisfaction research, making it ideal for businesses seeking to understand product perception and improve customer experience.

review_titleverified_purchasesentiment_label
Great Value!truepositive
Did not work as expectedtruenegative
It’s okayfalseneutral
review_idcustomer_idproduct_id+11 more
E-commerce12 cols · 200 rows

E-Commerce Reviews Sentiment Dataset

This dataset contains detailed e-commerce product reviews, including review text, star ratings, sentiment labels, and metadata such as helpful votes and verified purchase status. It is ideal for training and evaluating sentiment analysis models, understanding customer feedback, and powering recommendation engines. The rich structure supports both NLP research and practical business applications.

product_namereview_titlesentiment_label
Wireless Bluetooth EarbudsAmazing Sound Quality!positive
Stainless Steel Water BottleKeeps Drinks Cold For Hourspositive
Foldable Laptop StandMakes Working Easierpositive
review_idproduct_iduser_id+6 more
E-commerce13 cols · 200 rows

Product Reviews

This dataset provides detailed customer product reviews, including star ratings, review text, helpfulness votes, verified purchase status, and tracked responses from sellers or support teams. It enables comprehensive sentiment analysis, product feedback evaluation, and customer engagement tracking for e-commerce platforms.

review_titleverified_purchaseresponse_author
Absolutely fantastic!trueSupportTeam
Works as expectedtrue–
Not worth the pricetrueQA_Manager
review_idproduct_idcustomer_id+7 more
Social Media14 cols · 200 rows

Social Media Sentiment Dataset

This dataset contains synthetic social media posts and their associated comments, each labeled with sentiment (positive, negative, neutral) and optional sentiment scores. It is designed for training and benchmarking machine learning and NLP models in social analytics, enabling fine-grained sentiment analysis across multiple platforms and languages.

post_sentiment_labelcomment_sentiment_labelplatform
positivepositivetwitter
positivepositivefacebook
positivepositiveinstagram
post_iduser_idpost_datetime+8 more
Education13 cols · 200 rows

Student Feedback Sentiment Analysis

This dataset provides detailed student feedback on courses, enriched with sentiment analysis labels and scores, course and instructor metadata, and student academic context. It enables educators and administrators to systematically analyze curriculum effectiveness, identify areas for improvement, and track sentiment trends over time for data-driven decision making.

sentiment_labelcourse_titleinstructor_name
positiveIntroduction to Computer ScienceDr. Alicia Thornton
neutralPrincipios de EconomíaProf. Ricardo Alvarez
neutral数据结构基础Dr. Li Wei
feedback_idstudent_idcourse_id+7 more
Finance13 cols · 200 rows

Financial News Sentiment Streams

This dataset aggregates real-time sentiment scores and metadata for financial news headlines, enabling rapid detection of market-moving events and trends. It includes headline text, publication details, sentiment analysis, relevance to financial markets, and links to affected stocks and sectors. Ideal for quantitative trading, risk monitoring, and financial news analytics.

source_namesentiment_labelevent_type
Reuterspositiveearnings
Handelsblattneutralmacroeconomic
CNBCpositiveearnings
headline_idheadline_textpublished_datetime+7 more
Healthcare11 cols · 200 rows

Healthcare Chatbot Intent Dataset

This dataset provides detailed, synthetic healthcare chatbot conversations with annotated intent labels, message sequencing, and extracted entities. Designed for training and evaluating conversational AI, it supports intent classification, dialogue modeling, and entity recognition in healthcare virtual assistants. The dataset enables robust analysis of user-bot interactions for improved patient engagement and automation.

sender_typeintent_labelconversation_topic
userbook_appointmentappointments
botrequest_timeappointments
userprovide_timeappointments
transcript_idmessage_idmessage_order+5 more
Technology12 cols · 200 rows

User Feedback Text Analysis Set

This dataset contains synthetic, multi-domain user feedback entries, each labeled for sentiment, theme, and actionability to support machine learning applications in automated response systems and product improvement. It enables granular analysis of user concerns, trends, and actionable insights across different product areas, making it ideal for customer experience optimization and workflow automation.

sentiment_labeltheme_labelactionability_label
positivefeature_requestactionable
negativeperformanceactionable
mixedsupportneeds_clarification
feedback_iduser_iddomain+6 more
Technology20 cols · 200 rows

IT Service Ticket Classification

This dataset contains detailed records of IT service tickets, combining structured metadata (such as priority, category, and assignment) with rich ticket descriptions suitable for natural language processing. It enables automated ticket triage, prioritization, and advanced analytics for IT support operations, making it ideal for machine learning and process optimization.

ticket_subjectcategorypriority
Laptop battery fails to chargehardwaremedium
Cannot print from office printerhardwarehigh
Server room temperature alerthardwarecritical
departmentstatusurgency+14 more

Questions

Are the texts written by real people?
No. They are synthetic, so they contain no personal data and can be published or shared in class.
Can I get more examples of one class?
Yes. Open the dataset in GoMask Data Factory and ask for the balance you need, then generate up to a million rows.
What format is best for NLP work?
JSONL or CSV. Both load straight into pandas or Hugging Face datasets.

What should your data show?

Preview 20 rows free
No signup. No card.