Industrial wireless sensor selection method by using decision tree

Abstract

Presently, an industrial wireless sensor (IWS) has been interested and widely used by various industries. There are many industries trying to adopt qualified IWS products in competitive market for gaining advantage. The challenge in selection IWS includes not only the price, but also several factors, e.g., data rate, output power, operating voltage, current transmitting, current receiving, operate temperature, brand and so on. These factors cause difficulty for making decision by engineers or project managers, since there are a number of factors to select simultaneously in order to find the optimal IWS to use. In this paper, we propose the method for IWS selection by using decision tree in data mining. The proposed method is suitable for the problems involving a multiple factors selection simultaneously, as in the IWS selection problem. The IWS factors used for attribute decision are defined in this work, as well as the model of training set preferences. The classification method and decision tree technique are also derived in the paper to acquire the result, including the cross validation check for the result obtained. The result expressed that the decision tree provides the correct decision to optimal classification. It is rationality and feasibility for selection the qualified IWS products.