Dutch Social

The Dutch Social Dataset is a collection of tweets primarily written in Dutch. Let me provide you with some details about this dataset:

• Size: It contains approximately 271,342 tweets.
• Language: The tweets are filtered to include only those in the Dutch language or by users who have specified their location information within the Netherlands geographical boundaries.
• Creators: The annotations were generated by machine learning models, and the source datasets are original.
• License: The dataset is available under the cc-by-nc-4.0 license.
• Content: Each tweet includes information such as the full text, translation, screen name, description, location, and other relevant details.
• Geolocation: Some tweets also provide geolocation information (latitude, longitude, and altitude) for specific points.
• Sentiment: Sentiment patterns are included, ranging from negative to positive.
• Labels: The dataset includes class labels, such as "Maar, er iets nuttigs mee doen? Zie jij 'm vertrekken?" (However, there is something useful to do with it? Do you see him leave?) ¹.