Périodique
Squashing theory : a prediction approach for drug behaviour
("Squashing Theory" : une approche prédictive du comportement toxicomaniaque)
Auteur(s) :
BUSCEMA, M.
Année
1995
Langue(s) :
Anglais
ISBN :
8756-8233
Refs biblio. :
10
Domaine :
Drogues illicites / Illicit drugs
Note générale :
Drugs and Society, 1995, 8, (3-4), 103-110
Résumé :
FRANÇAIS :
Cet article décrit la "Squashing Theory" appliquée à la prévision du comportement d'addiction. Les résultats obtenus à partir du modèle informatique, qui prend en compte de nombreuses données d'ordre biologique, psychologique et social, se sont révélés assez proches de ceux observés sur des cas réels.
ENGLISH :
This paper describes a prediction approach called Squashing Theory. Squashing Theory, as used in this example to predict drug behavior, incorporates biological, psychological and sociological measures. The Artificial Feed Forward Neural Network - a recently developed computer architecture inspired by the brain's structure (Dayhoff, 1990) - is the framework for Squashing Theory. The Network was computer programmed by the Semeion Research Center in Rome, Italy. The model was able to predict drug behavior at the 92% level on prototypical cases and at the 80% level on uncertain cases based on self reported drug use from two norming samples on the prediction sample. (Author' s abstract)
Cet article décrit la "Squashing Theory" appliquée à la prévision du comportement d'addiction. Les résultats obtenus à partir du modèle informatique, qui prend en compte de nombreuses données d'ordre biologique, psychologique et social, se sont révélés assez proches de ceux observés sur des cas réels.
ENGLISH :
This paper describes a prediction approach called Squashing Theory. Squashing Theory, as used in this example to predict drug behavior, incorporates biological, psychological and sociological measures. The Artificial Feed Forward Neural Network - a recently developed computer architecture inspired by the brain's structure (Dayhoff, 1990) - is the framework for Squashing Theory. The Network was computer programmed by the Semeion Research Center in Rome, Italy. The model was able to predict drug behavior at the 92% level on prototypical cases and at the 80% level on uncertain cases based on self reported drug use from two norming samples on the prediction sample. (Author' s abstract)
Affiliation :
Univ. Kentucky, Lexington, KY 40506
Etats-Unis. United States.
Etats-Unis. United States.
Historique