Abstract

The method of discriminant analysis is one of the multivariate statistical methods that used to deal with descriptive data, and it depends on building a discrimination function , which is a linear combination of a set of independent variables, and this function works to reduce the similarity in classification errors. Discriminate analysis is used to classify one or more items into correct groups with the lowest possible classification error. The linear discriminate function was introduced by (Fisher) in (1936), who suggested that the classification should be based on a linear combination of the discriminating variables. By maximizing group differences on the one hand and minimizing differences within groups on the other hand, this method is suitable when the classification error is as low as possible. Work was done on the empirical work method, in which we begin to observe the problem, then hypotheses were made for it then tested, that is, by selecting a random sample and generalizing its results to the community. Determine the factors affecting the incidence of lung cancer and bone cancer and classifying the data using the linear discrimination function in order to reach the best classification for some types of cancerous tumors on the basis of the probability criterion of the least possible classification error. The most important factor affecting the incidence of lung and bone cancer is the gender of the patient, then the period of stay in the hospital, and finally the work of the patient