论文信息 - Intelligent Building Analysis Based on EM Algorithm in Generalized Linear Model with Multi-dimensional Cyclic Matrix

Intelligent Building Analysis Based on EM Algorithm in Generalized Linear Model with Multi-dimensional Cyclic Matrix

Evaluation method is a key part in the analysis of intelligent building features. In this paper, we propose an index system with 35 indexes to evaluate the architecture. Then we construct a generalized linear regression model on the condition that both dependent variables and regression coefficients can be estimated. Moreover we calculate the maximum likelihood of parameters by using EM algorithm and recurring circulation matrix to avoid the summations and circulations. Our algorithm can generate satisfactory weight of indexes.

Hengqing Tong | Qian Ding | Yajie Cheng

[1] John A. Bernaden,et al. The Intelligent Building Sourcebook , 1988 .

[2] Wei Pan,et al. Extending the Iterative Convex Minorant Algorithm to the Cox Model for Interval Censored Data , 2011 .

[3] Hengqing Tong,et al. Evaluation model and its iterative algorithm by alternating projection , 1993 .

[4] Xiao-Li Meng,et al. On the rate of convergence of the ECM algorithm , 1994 .

[5] Li Chun-lei. The Research on Synthetic Evaluation Target System of Optimum Intelligent Building Design , 2003 .

[6] Z Yang. APPLICATIONS OF THE GENERALIZED MODEL OF LEAST SQUARES , 1982 .

[7] Masahiro Kurodaa,et al. Accelerating the convergence of the EM algorithm using the vector algorithm , 2006 .

[8] Charles J. Geyer,et al. Constrained Maximum Likelihood Exemplified by Isotonic Convex Logistic Regression , 1991 .

[9] Jinwen Ma,et al. On the correct convergence of the EM algorithm for Gaussian mixtures , 2005, Pattern Recognit..

[10] Wu Cheng. A NEW EVALUATING METHOD OF INTELLIGENT BUILDING PERFORMANCE , 2002 .