Mixed datasets consist of both numeric and categorical attributes. Various k-means-based clustering algorithms have been developed for these datasets. Generally, these algorithms use random partition as a starting point, which tends to produce different clustering results for different runs. In this paper, we propose, initKmix, a novel algorithm for finding an initial partition for k-means-based clustering algorithms for mixed datasets. In the initKmix algorithm, a k-means-based clustering algorithm is run many times, and in each run, one of the attributes is used to create initial clusters for that run. The clustering results of various runs are combined to produce the initial partition. This initial partition is then used as a seed to a k-means-based clustering algorithm to cluster mixed data. Experiments with various categorical and mixed datasets showed that initKmix produced accurate and consistent results, and outperformed the random initial partition method and other state-of-the-art initialization methods. Experiments also showed that k-means-based clustering for mixed datasets with initKmix performed similar to or better than many state-of-the-art clustering algorithms for categorical and mixed datasets.

In one-class classification problems, only the data for the target class is available, whereas the data for the non-target class may be completely absent. In this paper, we study one-class nearest neighbor (OCNN) classifiers and their different variants. We present a theoretical analysis to show the relationships among different variants of OCNN that may use different neighbors or thresholds to identify unseen examples of the non-target class. We also present a method based on inter-quartile range for optimizing parameters used in OCNN in the absence of non-target data during training. Then, we propose two ensemble approaches based on random subspace and random projection methods to create accurate OCNN ensembles. We tested the proposed methods on 15 benchmark and real world domain-specific datasets and show that random-projection ensembles of OCNN perform best.

A classifier ensemble is a combination of diverse and accurate classifiers. Generally, a classifier ensemble performs better than any single classifier in the ensemble. Naive Bayes classifiers are simple but popular classifiers for many applications. As it is difficult to create diverse naive Bayes classifiers, naive Bayes ensembles are not very successful. In this paper, we propose Random Subclasses (RS) ensembles for Naive Bayes classifiers. In the proposed method, new subclasses for each class are created by using 1-Nearest Neighbor (1-NN) framework that uses randomly selected points from the training data. A classifier considers each subclass as a class of its own. As the method to create subclasses is random, diverse datasets are generated. Each classifier in an ensemble learns on one dataset from the pool of diverse datasets. Diverse training datasets ensure diverse classifiers in the ensemble. New subclasses create easy to learn decision boundaries that in turn create accurate naive Bayes classifiers. We developed two variants of RS, in the first variant RS(2), two subclasses per class were created whereas in the second variant RS(4), four subclasses per class were created. We studied the performance of these methods against other popular ensemble methods by using naive Bayes as the base classifier. RS(4) outperformed other popular ensemble methods. A detailed study was carried out to understand the behavior of RS ensembles.

In this paper, we present a novel ensemble method random projection random discretization ensembles(RPRDE) to create ensembles of linear multivariate decision trees by using a univariate decision tree algorithm. The present method combines the better computational complexity of a univariate decision tree algorithm with the better representational power of linear multivariate decision trees. We develop random discretization (RD) method that creates random discretized features from continuous features. Random projection (RP) is used to create new features that are linear combinations of original features. A new dataset is created by augmenting discretized features (created by using RD) with features created by using RP. Each decision tree of a RPRD ensemble is trained on one dataset from the pool of these datasets by using a univariate decision tree algorithm. As these multivariate decision trees (because of features created by RP) have more representational power than univariate decision trees, we expect accurate decision trees in the ensemble. Diverse training datasets ensure diverse decision trees in the ensemble. We study the performance of RPRDE against other popular ensemble techniques using C4.5 tree as the base classifier. RPRDE matches or outperforms other popular ensemble methods. Experiments results also suggest that the proposed method is quite robust to the class noise.