There are many prior works of modeling travel behaviors. Most of them are investigated under the assumption that many kinds of data such as that of Person Trip (PT), which surveys travel behaviors, are available. Ther...
详细信息
ISBN:
(纸本)9783642045912
There are many prior works of modeling travel behaviors. Most of them are investigated under the assumption that many kinds of data such as that of Person Trip (PT), which surveys travel behaviors, are available. Therefore, they do not, consider an application to cities where the survey is not examined. In this paper, we propose a method for estimating travel behaviors rising zone characteristics which is obtained from structural data, of city. Focusing on dependent relationships between travel behaviors and city structure, we estimate the travel behaviors by means of the relationships. We first define trip and zone characteristics, and then introduce our method. With our method, we make use of Bayesian network constructed with PT data and the structural data. In addition, we show the effectiveness of our method through evaluation experiments.
Customer losing problems are concerned by telecom operators as market becoming more competitive. Based on data mining technology, Bayesian networks classifier is used in the analysis of the problems. During the proces...
详细信息
Customer losing problems are concerned by telecom operators as market becoming more competitive. Based on data mining technology, Bayesian networks classifier is used in the analysis of the problems. During the process of Bayesian networks modeling, k2 and MCMC algorithms are utilized together. Effective variables are distilled through topology of networks, and churn rules are drawn based on CPT (condition probability table), then high probability churn customer groups are obtained. Considering loss function in classifier, different criterions and their class effects are provided. In contrast with other algorithm, such as decision tree and ANN (artificial neural networks), Bayesian networks can be modeled without over-sampling, when churn rate is relatively low.
The structure and parameters of a belief network are learned in order to classify images enabling the detection of genetic abnormalities. We compare a structure learned from the data to another structure obtained util...
详细信息
The structure and parameters of a belief network are learned in order to classify images enabling the detection of genetic abnormalities. We compare a structure learned from the data to another structure obtained utilizing expert knowledge and to the naive Bayesian classifier and study quantization in comparison to density estimation in parameter learning. (C) 2004 Elsevier B.V. All rights reserved.
暂无评论