Title :
Artificial Intelligence for Ethology: Neural Networks in the Prediction of Human Behavior
Author :
Koutsomanis, Demetrios
Author_Institution :
Lab. de Biosociologie Animale et Humaine, Univ. Paris V, La Sorbonne, Paris, France
Abstract :
Can we reliably predict human behavior using Artificial Intelligence? Can traditional methods of Ethology, such as an ethogramm, be implted to achieve this goal ? This is the aim of this study which used observational data collected from ambulatory addictology patients of a district general hospital, using traditional methodological tools of Ethology (ethograms). Observed versus predicted data after adequate training of an appropriately built neural network correlated strongly (Pearson two tailed) and did not differ significantly (t-test). We conclude that Neural Networks may, under defined methodological circumstances, reliably predict human behavior.
Keywords :
artificial intelligence; behavioural sciences computing; hospitals; ambulatory addictology patients; artificial intelligence; district general hospital; ethogram; ethology; human behavior prediction; neural networks; Accuracy; Artificial intelligence; Artificial neural networks; Computational modeling; Psychology; Training; Artificial Intelligence; Human Behavior; Neural Networks;
Conference_Titel :
Computer Modelling and Simulation (UKSim), 2014 UKSim-AMSS 16th International Conference on
Print_ISBN :
978-1-4799-4923-6
DOI :
10.1109/UKSim.2014.106