All tasks consist of the detection of mental disorders in users based on their comments posted on Telegram and Twitch. Given a history of messages about a user, the goal is to identify whether the user suffers from the disorder or not and the context that influences the mental health problem.
This is a binary classification task aimed at determining whether a user is at high risk (label = 1) or low risk (label = 0) of developing a gambling-related disorder based on their messages. The objective is to enable early detection and facilitate timely interventions.
This task is similar to Task 1 but with an additional requirement: As all users are at risk (low or high), no matter the decision, for all of them the model should identify the specific type of addiction associated with the disorder.The available labels are Betting, Online Gaming, Trading, and Lootboxes. Below are some examples:
Betting: Gambling activity where individuals place bets on sports-related events, aiming to win money based on the outcome. This is particularly popular among young people, especially in football betting.
User1 (messages):
“yo de frees no hable“
"pues de pago si hay buenos , pero a mi me ha llevado mi tiempo emplea tu el tuyo”
Online Gaming: Gambling activity where individuals participate in traditional games of chance, such as roulette, blackjack, or slot machines, wagering money either in person or online
User2 (messages):
“24 céntimos con los free spins , Roma no se construyó en un dia"
"pues de pago si hay buenos , pero a mi me ha llevado mi tiempo emplea tu el tuyo”
Trading and crypto: Gambling activity where individuals engage in speculative investments, particularly in cryptocurrencies, facing uncertainty regarding financial gains or losses.
User3 (messages):
“Siii fue cuestión de rapidez porqje subio a 60 y rapidio se desinflo“
"interesante”
Lootboxes: Gambling activity where individuals purchase virtual items in video games that contain randomized content, often using real-world money, introducing an element of financial risk.
User4 (messages):
“y al que le toco sirve para intercambio“
"para otra cosa no”
Notes:
No matter whether the subject is predicted as low or high risk, it is always associated with a type of gambling problem.
The prediction of the type of addiction to be considered for evaluation will be the one received in the last round.
All users belong to at least 1 of the categories and at most 1 of the categories.