Modeling and Prediction of Human Behavior

Article Properties
  • Language
    English
  • Publication Date
    1999/01/01
  • Indian UGC (Journal)
  • Refrences
    7
  • Citations
    212
  • Alex Pentland Massachusetts Institute of Technology, Cambridge, MA 02139, U.S.A.
  • Andrew Liu Nissan Cambridge Basic Research, Cambridge, MA 02142, U.S.A.
Abstract
Cite
Pentland, Alex, and Andrew Liu. “Modeling and Prediction of Human Behavior”. Neural Computation, vol. 11, no. 1, 1999, pp. 229-42, https://doi.org/10.1162/089976699300016890.
Pentland, A., & Liu, A. (1999). Modeling and Prediction of Human Behavior. Neural Computation, 11(1), 229-242. https://doi.org/10.1162/089976699300016890
Pentland A, Liu A. Modeling and Prediction of Human Behavior. Neural Computation. 1999;11(1):229-42.
Journal Categories
Medicine
Internal medicine
Neurosciences
Biological psychiatry
Neuropsychiatry
Science
Mathematics
Instruments and machines
Electronic computers
Computer science
Technology
Electrical engineering
Electronics
Nuclear engineering
Electronics
Technology
Mechanical engineering and machinery
Description

Can we predict what someone will do next? This research presents a novel approach to modeling and predicting human behavior by representing it as a sequence of dynamic models, specifically Kalman filters, linked together by a Markov chain. This approach aims to capture the dynamic nature of human actions and provide accurate predictions over short time horizons. The core idea is that complex human behaviors can be broken down into a series of simpler, dynamic models that capture the underlying patterns. The Markov chain then governs the transitions between these models, allowing the system to adapt to changing circumstances. This framework is tested in an experiment involving automobile drivers, where the model achieves impressive accuracy in predicting subsequent actions based on initial preparatory movements. This modeling paradigm offers a powerful tool for understanding and anticipating human behavior in various contexts. The implications of this research extend to areas such as robotics, artificial intelligence, and human-computer interaction, where predicting human actions is crucial for seamless collaboration.

Published in Neural Computation, this article is suitable for the journal's audience due to its intersection of computational modeling and neuroscience. The use of dynamic models and Markov chains to represent human behavior aligns with the journal's focus on computational approaches to understanding brain function and behavior.

Refrences
Citations
Citations Analysis
The first research to cite this article was titled Looking at people: sensing for ubiquitous and wearable computing and was published in 2000. The most recent citation comes from a 2024 study titled Looking at people: sensing for ubiquitous and wearable computing . This article reached its peak citation in 2019 , with 24 citations.It has been cited in 135 different journals, 14% of which are open access. Among related journals, the IEEE Transactions on Intelligent Transportation Systems cited this research the most, with 18 citations. The chart below illustrates the annual citation trends for this article.
Citations used this article by year