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.

  • opened 18 times
  • last updated 21 Aug 2025
  • by GoMask
The brief that made it

Training sentiment analysis models for social media platforms

Sample rows

preview · 8 of 200 rows · all 14 columns
post_idstringcomment_sentiment_labelstringcomment_sentiment_scorefloatplatformstringuser_idstringpost_datetimedatetimepost_sentiment_labelstringpost_sentiment_scorefloatcomment_idstringcomment_user_idstringcomment_datetimedatetimecomment_contentstringlanguagestringpost_contentstring
post_0001positive0.93twitteruser_00012024-01-01T00:05:00Zpositive0.85comm_0001user_00212024-01-01T00:07:00ZThank you! Hope you have a wonderful year too.enHappy New Year! Wishing everyone a joyful 2024.
post_0002positive0.81facebookuser_00022024-02-29T08:12:00Zpositive0.74comm_0002user_00182024-02-29T08:14:00ZEnjoy the extra day! Make it count.enLeap Day is here! Time for something special.
post_0003positive0.92instagramuser_00032024-03-05T15:20:00Zpositive0.88comm_0003user_00222024-03-05T15:22:00ZCongrats! That's an amazing achievement.enJust finished my first marathon. Feeling accomplished!
post_0004positive0.67reddituser_00042024-04-12T10:03:00Znegative-0.69comm_0004user_00192024-04-12T10:05:00ZHang in there! Tomorrow's a new day.enHad a tough day at work. Wish things would improve.
post_0005positive0.81twitteruser_00052024-05-01T09:41:00Zpositive0.78comm_0005user_00202024-05-01T09:44:00Z¡Claro que sí! Ánimo y que sea genial.es¡Buenos días a todos! Hoy será un buen día.
post_0006positive0.79instagramuser_00062024-06-15T18:20:00Zpositive0.76comm_0006user_00232024-06-15T18:24:00ZBest time of the year! Enjoy it.enSummer vibes. Loving the sunshine and fresh air.
post_0007positive0.85facebookuser_00072024-07-04T11:30:00Zpositive0.91comm_0007user_00242024-07-04T11:31:00ZCan't wait for the celebrations!enHappy Independence Day! Fireworks tonight.
post_0008positive0.94instagramuser_00082023-12-24T23:59:00Zpositive0.96comm_0008user_00252023-12-25T00:02:00ZWishing you and yours a joyful holiday!enMerry Christmas Eve! Family time is the best.

What the 200 rows show

from the 200-row sample

Neutral (comment sentiment label) stands out: mean comment_sentiment_score is 0.05, against 0.76 for the rest.

  • 0.74median comment_sentiment_score
  • 3post sentiment labels
  • 5languages
  • 0.73median post_sentiment_score
Mean comment_sentiment_score by comment_sentiment_label176 rows
00.40.80.76positive143 rows0.05neutral33 rows
comment_sentiment_score176 rows, in bands of 0.2
040801320107657-0.20.41comment_sentiment_score →

Median 0.74, from -0.02 to 0.96.

platform200 rows · 5 values
  1. facebook52
  2. twitter49
  3. instagram43
  4. other30
  5. reddit26
14 columns by typefrom the column list below
  • string 10
  • float 2
  • datetime 2

Columns

14 columns in three groups
blueprint · 14 columns
columntypedescriptionexample
Text 10 columns
post_idstringUnique identifier for each social media postuniquepost_0001
user_idstringUnique identifier for the user who created the postuser_0001
post_contentstringFull textual content of the social media post1 or moreA
post_sentiment_labelstringSentiment label assigned to the postpositive · negative · neutralpositive
comment_idstringUnique identifier for each comment on a postunique · optionalcomm_0001
comment_user_idstringUnique identifier for the user who made the commentoptionaluser_0021
comment_contentstringFull textual content of the comment1 or more · optionalGenieß die frische Luft!
comment_sentiment_labelstringSentiment label assigned to the commentpositive · negative · neutral · optionalpositive
platformstringSocial media platform where the post/comment was madetwitter · facebook · instagram · reddit · other · optionaltwitter
languagestringLanguage of the post or comment (ISO 639-1 code)5 languages · optionalen
Numbers 2 columns
post_sentiment_scorefloatNumerical sentiment score for the post (e.g., -1 to 1)-1 to 1 · optional0.85
comment_sentiment_scorefloatNumerical sentiment score for the comment (e.g., -1 to 1)-1 to 1 · optional0.93
Dates and times 2 columns
post_datetimedatetimeTimestamp when the post was created2024-01-01T00:05:00Z
comment_datetimedatetimeTimestamp when the comment was createdoptional2024-01-01T00:07:00Z

15 SQL practice questions for the Social Media Sentiment Dataset

Written for this data and checked against it: 12 SQL and 3 Power BI exercises from beginner to advanced. Try the first three free; the practice kit has all 15, a starter .sql file and an answer key with every expected result.

  1. 01BeginnerFreeGROUP BY · COUNT

    Posts per platform

    Count the posts on each platform. Return the columns platform and post_count, ordered by post_count from highest to lowest, then by platform alphabetically.

    Tables: social_media_sentiment_dataset · your result: 2 columns, 5 rows
  2. 02BeginnerFreeDISTINCT · ORDER BY

    Distinct languages

    List each distinct language found in the data. Return a single column named language, sorted alphabetically.

    Tables: social_media_sentiment_dataset · your result: 1 columns, 5 rows
  3. 03BeginnerFreeWHERE · ORDER BY

    Strongly positive English posts

    Find English posts (language = 'en') with a post_sentiment_score above 0.9. Return post_id, platform and post_sentiment_score, ordered by post_sentiment_score from highest to lowest, then by post_id.

    Tables: social_media_sentiment_dataset · your result: 3 columns, 13 rows
  1. 04Average post score by labelBeginner
  2. 05Average post score by languagePower BI
  3. 06Platforms with many postsIntermediate
  4. 07Score bucketsIntermediate
  5. 08Posts per yearIntermediate
  6. 09Languages above overall averageIntermediate
  7. 10Comment coverage measurePower BI
  8. 11Rank posts within platformAdvanced
  9. 12Yearly change in post countAdvanced
  10. 13Running total of posts by monthAdvanced
  11. 14Platform share of postsAdvanced
  12. 15Anomalous comment scoresPower BI

Get the practice kit: the full dataset, all 15 exercises, a starter .sql and the answer key. One-time $3.99.

Get the practice kit · $3.99

Use it for

  • median comment…0.74176 rowsmean comment sentimen…0.76positi…0.05neutral

    A social media dashboard

    Comment_sentiment_score by comment_sentiment_label and a breakdown of platform. Excel, Power BI or Tableau.

  • Why do the 33 neutral rows have a mean comment_sentiment_score of 0.05?

    A root-cause class exercise

    Hand out the rows and one question. The answer is in the data, not in the brief.

  • A software demo

    Believable posts with user_id, post_datetime and post_content to fill a screen in front of a buyer.

Not quite right?

Make it yours.

Same 14 columns, your size and your rules. See 20 rows before you pay.

Preview 20 rows free

1,000 rows: $1.99 · 10,000: $12.99One-time. No subscription. All prices

This dataset200 rows14 columns
Yours1,000 rows14 columns

blueprint · social-media-sentiment-dataset

Behind this dataset

Same schema. As many rows as you need.

These 200 rows came out of a blueprint — 14 columns with generation rules behind each one. Open it in Data Factory to retune a column, add your own, wire in foreign keys, and run it at the size you actually need.

Rules it was built with
  • Usernames anonymized and randomized
  • Sentiment labeled: positive, negative, neutral
  • Post lengths vary 20-240 characters
  • Hashtags and emojis randomly included
Rows
Open the blueprint in Data Factory

1 credit per row. Preview 20 rows free. Buy rows for the full run.

Exports
CSV, JSON, JSONL, Parquet, SQL, Excel, TSV, XML
Licence
yours to use, including commercially
API slug
social-media-sentiment-dataset

What should your data show?

Preview 20 rows free
No signup. No card.