ترجمه مقاله نقش ضروری ارتباطات 6G با چشم انداز صنعت 4.0
- مبلغ: ۸۶,۰۰۰ تومان
ترجمه مقاله پایداری توسعه شهری، تعدیل ساختار صنعتی و کارایی کاربری زمین
- مبلغ: ۹۱,۰۰۰ تومان
Abstract
A probability parameter updating method for Bayesain network with fixed structure and initial probability parameters is discussed, when new data set is available. It should reserve the original probability parameters as much as possible and also reflect the probability distribution contained in data set furthest. Principle of parameters updating based on MLE is discussed, algorithms for complete data set and incomplete data set are put forward. A case study is carried out to show how the algorithm works.
VI. SUMMARY
The main goal of this paper is to propose a method to update BN’s probability parameters with fixed structure and original parameters when new data set is available, thus to retain the probability information of the original model and also reflect the probability distribution contained in the new data set. After the principle of parameters updating based on MLE discussed, algorithms for complete data set and incomplete data set are put forward. A case study is carried out at last and learning effects by different learning speeds are discussed also.