A Comparison of Projection Pursuit and Neural Network Regression Modeling
نویسندگان
چکیده
Two projection based feedforward network learning methods for modelfree regression problems are studied and compared in this paper: one is the popular back-propagation learning (BPL); the other is the projection pursuit learning (PPL). Unlike the totally parametric BPL method, the PPL non-parametrically estimates unknown nonlinear functions sequentially (neuron-by-neuron and layer-by-Iayer) at each iteration while jointly estimating the interconnection weights. In terms of learning efficiency, both methods have comparable training speed when based on a GaussNewton optimization algorithm while the PPL is more parsimonious. In terms of learning robustness toward noise outliers, the BPL is more sensitive to the outliers.
منابع مشابه
Performance comparison of land change modeling techniques for land use projection of arid watersheds
The change of land use/land cover has been known as an imperative force in environmental alteration, especially in arid and semi-arid areas. This research was mainly aimed to assess the validity of two major types of land change modeling techniques via a three dimensional approach in Birjand urban watershed located in an arid climatic region of Iran. Thus, a Markovian approach based on two suit...
متن کاملImplementing projection pursuit learning
This paper examines the implementation of projection pursuit regression (PPR) in the context of machine learning and neural networks. We propose a parametric PPR with direct training which achieves improved training speed and accuracy when compared with nonparametric PPR. Analysis and simulations are done for heuristics to choose good initial projection directions. A comparison of a projection ...
متن کاملComparison of artificial neural network with logistic regression in prediction of tendency to surgical intervention in nurses
Introduction: Logistic regression is one of the modeling methods for bipartite dependent variables. On the other hand, artificial neural network is a flexible method with the least limitation. The importance of growing unnecessary beauty surgeries and the importance of prediction and classification made us consider the present study, with the aim of comparing logistic regression and artificial ...
متن کاملP/E Modeling and Prediction of Firms Listed on the Tehran Stock Exchange; a New Approach to Harmony Search Algorithm and Neural Network Hybridization
Investors and other contributors to stock exchange need a variety of tools, measures, and information in order to make decisions. One of the most common tools and criteria of decision makers is price-to earnings per share ratio. As a result, investors are in pursuit of ways to have a better assessment and forecast of price and dividends and get the highest returns on their investment. Previous ...
متن کاملApplication of Linear Regression and Artificial NeuralNetwork for Broiler Chicken Growth Performance Prediction
This study was conducted to investigate the prediction of growth performance using linear regression and artificial neural network (ANN) in broiler chicken. Artificial neural networks (ANNs) are powerful tools for modeling systems in a wide range of applications. The ANN model with a back propagation algorithm successfully learned the relationship between the inputs of metabolizable energy (kca...
متن کامل