Mental Health Twitter Dataset for Sentiment Analysis (Jan–May 2025)

This dataset contains 10,699 English-language tweets collected from Twitter (X) using keywords related to mental health: "depression", "anxiety", "stress", and "mental health". The data was gathered from January to May 2025 using the Twitter API. Each entry includes metadata such as tweet text, timestamps, engagement metrics (likes, replies, retweets), and media links (if available). The dataset is intended for use in Natural Language Processing research on sentiment classification, especially in the domain of mental health detection using transformer models like BERT and RoBERTa. Missing data is present in fields such as location and username, which have been anonymized or stripped for ethical and privacy compliance. Dataset Overview Total Rows (Entries): 10,699 Total Columns: 15 Language Focus: Mostly English (lang = en in 10,687 out of 10,699 rows) Content: The dataset contains tweets related to mental health topics such as anxiety, depression, and stress, collected from Twitter (X) between January and May 2025. Main Columns & Descriptions Column Name Description conversation_id_str Unique conversation/thread identifier created_at Timestamp of the tweet favorite_count Number of likes/favorites full_text The actual tweet text content id_str Tweet ID image_url URL to any attached image (if any) in_reply_to_screen_name Username if the tweet is a reply lang Language of the tweet (mostly en) location Tweet author's location (all null in this dataset) quote_count Number of times the tweet was quoted reply_count Number of replies retweet_count Number of retweets tweet_url Direct URL to the tweet user_id_str ID of the user who posted the tweet username Username of the author (all null in this dataset) Data Highlights Tweets with images: 2,999 rows contain a non-null image_url Replies: About 4,725 tweets are replies (in_reply_to_screen_name not null) Engagement: favorite_count: ranges from 0 to 34,364 retweet_count: ranges from 0 to 1,958 Most tweets have low engagement (median values are 0) Top mentioned account: elonmusk (mentioned in replies 61 times) Missing Data: location and username: completely null image_url: ~72% of tweets don’t include images