An AI tool to combat substance abuse


The USC Centre for Artificial Intelligence Researchers has developed a novel algorithm-based on Artificial Intelligence(AI) that uses social networks to sorts the substance abuse treatment intervention groups for the homeless youth. The algorithm is based on the participants, who volunteered for the programme, distributed into subgroups, or smaller groups, in a way that maintains healthy social relationships and demolish the social relations which are harmful to recovery.

"In order to improve the effectiveness of the intervention, you need to know how people will influence each other in a group.As we all know that substance abuse is highly affected by the influence of others," said Aida Rahmattalabi, a USC computer science graduate student and lead author of the study.

The process of helping the homeless youth if not went well it can lead to connections based on antisocial doing.

This problem is tackled by a team by seeing through AI perspective, creating an algorithm that keeps the information about how the individuals in a subgroup are connected and their prior history of drug use.

"Based on the information we have an influence model that explains how likely it is for an individual to adopt negative behaviors or change negative behaviors based on their participation in the group", Rahmattalabi noted.

"This helps us predict what happens when we group into smaller groups," she said.

The computational model of intervention was built based on the information gathered from the volunteered homeless youths, behavior theories, and observations of previous interventions. The analysis of the intervention groups shows that sometimes conducting the intervention could have a harmful effect on the group.

"In some cases, we found it's actually a bad idea to conduct the intervention. For example, if you a risk people in a group, it's better to not connect them with low-risk people."Says Rahamattalabi.

Read more news:

AI models can help unravel the mystery of the human brain

Ransomware payments affected South Koreans the most