A Retraining Procedure Application for Data Prediction

نویسندگان

  • I. NASTAC
  • A. COSTEA
چکیده

Articolul prezintă o arhitectură specială de reţea neuronală feedforward ce utilizează algoritmul de reantrenare pentru predicţia variabilelor de proces specifice unei aplicaţii industriale. Capacitatea modelului de extragere a unor informaţii relevante conferă un cadru util prentru reprezentarea relaţiilor existente în seriile temporale utilizate. Scopul principal este de a stabili un optim al arhitecturii şi al vectorilor de întârziere potriviţi pentru această predicţie de date. Prin evaluarea erorii la ieşire, după reantrenare, se evidenţiază faptul că această procedură poate înbunătăţi rezultatele modelului.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Retrainable Neural Networks for Image Analysis and Classification

A novel approach is presented in this paper for improving the performance of neural network classifiers in image recognition, segmentation or coding applications, based on a retraining procedure at the user level. The procedure includes a maximum a posteriori (MAP) estimation technique for optimally selecting a retraining data set from the image applied to the network during real life operation...

متن کامل

PREDICTING URBAN TRIP GENERATION USING A FUZZY EXPERT SYSTEM

One of the most important stages in the urban transportation planning procedure is predicting the rate of trips generated by each trac zone. Currently, multiple linear regression models are frequently used as a prediction tool. This method predicts the number of trips produced from, or attracted to each trac zone according to the values of independent variables for that zone. One of the main li...

متن کامل

Application of Artificial Neural Network and Fuzzy Inference System in Prediction of Breaking Wave Characteristics

Wave height as well as water depth at the breaking point are two basic parameters which are necessary for studying coastal processes. In this study, the application of soft computing-based methods such as artificial neural network (ANN), fuzzy inference system (FIS), adaptive neuro fuzzy inference system (ANFIS) and semi-empirical models for prediction of these parameters are investigated. Th...

متن کامل

Neural Networks Retraining in Image Analysis Problems

A novel approach is presented in this paper for improving the performance of neural network classifiers in image recognition, segmentation or coding applications, based on a retraining procedure at the user level. The procedure includes (a) a training algorithm for adapting the network weights to the current condition, (b) a maximum a posteriori (MAP) estimation procedure for optimally selectin...

متن کامل

Application of Grey System Theory in Rainfall Estimation

Considering the fact that Iran is situated in an arid and semi-arid region, rainfall prediction for the management of water resources is very important and necessary. Researchers have proposed various prediction methods that have been utilized in such areas as water and meteorology, especially water resources management. The present study aimed at predicting rainfall amounts using Grey Predicti...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2011