This work presents a classification algorithm based on genetic algorithms (GAs) that discovers comprehensible IF-THEN rules, in the spirit of data mining. The proposed GA has a flexible chromosome encoding where each chromosome corresponds to a classification rule. Although the number of genes (genotype) is fixed, the number of rule conditions (phenotype) is variable. The GA also has specific mutation operators for this chromosome encoding. The algorithm was evaluated on two public domain, realworld data sets (on the medical domains of dermatology and breast cancer). 1 Introduction This work presents a system based on genetic algorithms (GAs) to perform the task of classification. The system is evaluated in two medical domains: diagnosis of dermatological diseases and prediction of recurrence of breast cancer. The use of GAs in classification is an attempt to effectively exploit the large search space usually associated with classification tasks. The GA presented here was designed ac...

This paper describes a machine learning method, called Regression by Feature Projections (RFP), for predicting a real-valued target feature. In RFP training is based on simply storing the projections of the training instances on each feature separately. Prediction of the target value for a query point is obtained through two approximation procedures executed sequentially. The first approximation process is to find the individual predictions of features by using the K-nearest neighbor algorithm (KNN). The second approximation process combines the predictions of all features. During the first approximation step, each feature is associated with a weight in order to determine the prediction ability of the feature at the local query point. The weights, found for each local query point, are used in the second step and enforce the method to have an adaptive or context-sensitive nature. We have compared RFP with the KNN algorithm. Results on real data sets show that RFP is much faster than KNN, yet its prediction accuracy is comparable with the KNN algorithm.

A new instance-based learning method is presented for regression problems with high-dimensional data. As an instance-based approach, the conventional method, KNN, is very popular for classification. Although KNN performs well on classification tasks, it does not perform as well on regression problems. We have developed a new instance-based method, called Regression by Partitioning Feature Projections (RPFP) which is designed to meet the requirement for a lazy method that achieves high levels of accuracy on regression problems. RPFP gives better performance than well-known eager approaches found in machine learning and statistics such as MARS, rule-based regression, and regression tree induction systems. The most important property of RPFP is that it is a projectionbased approach that can handle interactions. We show that it outperforms existing eager or lazy approaches on many domains when there are many missing values in the training data.

In some domains, the cost of a wrong classification may be different for all pairs of predicted and actual classes. Also the benefit of a correct prediction is different for each class. In this paper, a new classification algorithm, called BCFP (for Benefit Maximizing Classifier on Feature Projections), is presented. The BCFP classifier learns a set of classification rules that will predict the class of a new instance with maximum benefit or minimum cost. BCFP represents a concept in the form of feature projections on each feature dimension separately. Classification in the BCFP algorithm is based on a voting among the individual predictions made on each feature. A genetic algorithm is used to select the relevant features. The performance of the BCFP algorithm is evaluated in terms of accuracy. As a case study, the BCFP algorithm is applied to the problem of diagnosis of gastric carcinoma. A lesion can be an indicator of one of nine different levels of gastric carcinoma. The benefit of correct classification of early levels is much more than that of late cases. Also, the cost of wrong classifications is different for all classes. Key Words Machine learning, feature projection, voting, benefit maximization 1.

Bankruptcy prediction has been an important decision-making process for financial analysts. One of the most common approaches for the bankruptcy prediction problem is the Discriminant Analysis. Also, the k-Nearest Neighbor classifier is very successful in such domains. This paper proposes a Feature Projection based classification algorithm, and explores its applicability to the problem of predicting bankruptcy of large firms. The algorithm is evaluated on a particular data set, and its performance is compared with the techniques mentioned above. The experiments indicate that the feature projection based classification algorithm introduced here performs better than these techniques.

Presence of irrelevant features is a fact of life in many realworld applications of classification learning. Although nearest-neighbor classification algorithms have emerged as a promising approach to machine learning tasks with their high predictive accuracy, they are adversely affected by the presence of such irrelevant features. In this paper, we describe a recently proposed classification algorithm called VFI5, which achieves comparable accuracy to nearest-neighbor classifiers while it is robust with respect to irrelevant features. The paper compares both the nearest-neighbor classifier and the VFI5 algorithms in the presence of irrelevant features on both artificially generated and real-world data sets selected from the UCI repository.

There is a great need for classification methods that can properly handle asymmetric cost and benefit constraints of classifications. In this study, we aim to emphasize the importance of classification benefits by means of a new classification algorithm, Benefit Maximizing classifier with Feature Intervals (BMFI) that uses feature projection based knowledge representation. Empirical results show that BMFI has promising performance compared to recent cost-sensitive algorithms in terms of benefit gained.

In this paper we present the application of a particular neuro-fuzzy system, named KERNEL, to the problem of differential diagnosis of erythematosquamous diseases, which represents a major problem in dermatology. A multistep learning strategy is adopted to obtain, starting directly from available data, a fuzzy rule base that can be used to identify the particular disease. The obtained classification results at the end of a two-phase experimental session are reported. Keywords: Neuro-fuzzy systems, medical application. 1

This paper is about the implementation of a visual tool for Differential Diagnosis of Erythemato-Squamous Diseases based on the