Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review
Comparison of Behavior Rate Models Based on Bayesian Belief Network. / Toropova, Aleksandra; Tulupyeva, Tatiana.
Studies in Systems, Decision and Control. Springer Nature, 2021. p. 510-521 (Studies in Systems, Decision and Control; Vol. 337).Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review
}
TY - CHAP
T1 - Comparison of Behavior Rate Models Based on Bayesian Belief Network
AU - Toropova, Aleksandra
AU - Tulupyeva, Tatiana
N1 - Publisher Copyright: © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2021
Y1 - 2021
N2 - We can use knowledge about human behavior frequency in many fields. There are many problems in sociology, psychology and other sciences studying human behavior where it becomes necessary to predict the behavior frequency. Direct evaluation of frequency is not always executable therefore developing of methods of indirect frequency evaluation is actual problem. The term behavior rate or frequency is determined as the mean number of behavior episodes happened during the particular period. A behavior episode is an activity happened in a certain time. Previously an approach based on Bayesian belief networks and data about last behavior episodes was presented. We suggest an improvement by including in the model a hidden variable representing an interval between the last episode in study period and the next episode, i.e. the first episode occurred after the study period. We compare this and the previous models. For learning and testing the models data are synthesized and collected from Vkontakte, the most popular social Network in Russia. The results can be used in various scientific fields, the subject of which is a human behavior: sociology, psychology, marketing etc.
AB - We can use knowledge about human behavior frequency in many fields. There are many problems in sociology, psychology and other sciences studying human behavior where it becomes necessary to predict the behavior frequency. Direct evaluation of frequency is not always executable therefore developing of methods of indirect frequency evaluation is actual problem. The term behavior rate or frequency is determined as the mean number of behavior episodes happened during the particular period. A behavior episode is an activity happened in a certain time. Previously an approach based on Bayesian belief networks and data about last behavior episodes was presented. We suggest an improvement by including in the model a hidden variable representing an interval between the last episode in study period and the next episode, i.e. the first episode occurred after the study period. We compare this and the previous models. For learning and testing the models data are synthesized and collected from Vkontakte, the most popular social Network in Russia. The results can be used in various scientific fields, the subject of which is a human behavior: sociology, psychology, marketing etc.
KW - Bayesian belief networks
KW - Behavior episodes
KW - Behavior frequency
KW - Behavior modeling
KW - Behavior rate
KW - Frequency estimation
KW - Hidden variables
KW - Social behavior
KW - Social behavior analysis
UR - http://www.scopus.com/inward/record.url?scp=85097257294&partnerID=8YFLogxK
UR - https://www.mendeley.com/catalogue/d60f40df-cabe-3088-b271-825254344940/
U2 - 10.1007/978-3-030-65283-8_42
DO - 10.1007/978-3-030-65283-8_42
M3 - Chapter
AN - SCOPUS:85097257294
T3 - Studies in Systems, Decision and Control
SP - 510
EP - 521
BT - Studies in Systems, Decision and Control
PB - Springer Nature
ER -
ID: 78033363