Annotated Tweets on Cannabis and Depression Association, 2017-2019

A sample of 9,888 tweets from January 2017 to February 2019 was collected using Cannabis and depression lexicons. The tweets were manually annotated by domain experts and research students into categories like 'Treat', 'Cause', and 'Addiction'. The project was sponsored by a National Institute on Drug Abuse grant to the Ohio Center of Excellence in Knowledge-enabled Computing.

Use Cases

• Relation extraction based on annotated categories linking cannabis use and depression.
• Entity disambiguation for terms related to cannabis products and depression symptoms mentioned in tweets.
• Relation prediction modeling to classify the perceived association between cannabis and depression in social media text.
• Public health research to understand public discourse on cannabis as a treatment or cause for depression.

Strengths

• 9,888 tweets manually annotated by domain experts, including substance use epidemiologists.
• Annotation scheme includes five specific relationship categories such as 'Treat' and 'Cause'.
• Data collection spans a defined time period from January 2017 to February 2019.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the full dataset is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Twitter, collected using Drug Abuse Ontology (DAO) terms and a depression lexicon.

Collection Method

Tweets were collected and a sample was manually annotated by domain experts and research students.

Time Range

January 2017 to February 2019