Search

Abstract The paper reports an experimentally identified list of benchmark data sets that are hard for representative classification and feature selection methods. This was done after systematically evaluating a total of 48 combinations of methods, involving eight state-of-the-art classification algorithms and six commonly used feature selection methods, on 129 data sets from the UCI repository (some data sets with known high classification accuracy were excluded). In this paper, a data set for classification is called hard if none of the 48 combinations can achieve an AUC over 0.8 and none of them can achieve an F-Measure value over 0.8; it is called easy otherwise. A total of 15 out of the 129 data sets were found to be hard in that sense. This paper also compares the performance of different methods, and it produces rankings of classification methods, separately on the hard data sets and on the easy data sets. This paper is the first to rank methods separately for hard data sets and for easy data sets. It turns out that the classifier rankings resulting from our experiments are somehow different from those in the literature and hence they offer new insights on method selection. It should be noted that the Random Forest method remains to be the best in all groups of experiments. [ABSTRACT FROM AUTHOR]

Personalization is the adaptation of the services to fit the user's interests, characteristics and needs. The key to effective personalization is user profiling. Apart from traditional collaborative and content-based approaches, a number of classification and clustering algorithms have been used to classify user related information to create user profiles. However, they are not able to achieve accurate user profiles. In this paper, we present a new clustering algorithm, namely Multi-Dimensional Clustering (MDC), to determine user profiling. The MDC is a version of the Instance-Based Learner (IBL) algorithm that assigns weights to feature values and considers these weights for the clustering. Three feature weight methods are proposed for the MDC and, all three, have been tested and evaluated. Simulations were conducted with using two sets of user profile datasets, which are the training (includes 10,000 instances) and test (includes 1000 instances) datasets. These datasets reflect each user's personal information, preferences and interests. Additional simulations and comparisons with existing weighted and non-weighted instance-based algorithms were carried out in order to demonstrate the performance of proposed algorithm. Experimental results using the user profile datasets demonstrate that the proposed algorithm has better clustering accuracy performance compared to other algorithms. This work is based on the doctoral thesis of the corresponding author. [ABSTRACT FROM AUTHOR]

Classification algorithms are the most commonly used data mining models that are widely used to extract valuable knowledge from huge amounts of data. The criteria used to evaluate the classifiers are mostly accuracy, computational complexity, robustness, scalability, integration, comprehensibility, stability, and interestingness. This study compares the classification of algorithm accuracies, speed (CPU time consumed) and robustness for various datasets and their implementation techniques. The data miner selects the model mainly with respect to classification accuracy; therefore, the performance of each classifier plays a crucial role for selection. Complexity is mostly dominated by the time required for classification. In terms of complexity, the CPU time consumed by each classifier is implied here. The study first discusses the application of certain classification models on multiple datasets in three stages: first, implementing the algorithms on original datasets; second, implementing the algorithms on the same datasets where continuous variables are discretised; and third, implementing the algorithms on the same datasets where principal component analysis is applied. The accuracies and the speed of the results are then compared. The relationship of dataset characteristics and implementation attributes between accuracy and CPU time is also examined and debated. Moreover, a regression model is introduced to show the correlating effect of dataset and implementation conditions on the classifier accuracy and CPU time. Finally, the study addresses the robustness of the classifiers, measured by repetitive experiments on both noisy and cleaned datasets. [ABSTRACT FROM AUTHOR]

Abstract: The rough set (RS) theory can be seen as a new mathematical approach to vagueness and is capable of discovering important facts hidden in that data. However, traditional rough set approach ignores that the desired reducts are not necessarily unique since several reducts could include the same value of the strength index. In addition, the current RS algorithms have the ability to generate a set of classification rules efficiently, but they cannot generate rules incrementally when new objects are given. Numerous studies of incremental approaches are not capable to deal with the problems of large database. Therefore, an incremental rule-extraction algorithm is proposed to solve these issues in this study. Using this algorithm, when a new object is added up to an information system, it is unnecessary to re-compute rule sets from the very beginning, which can quickly generate the complete but not repetitive rules. In the case study, the results show that the incremental issues of new data add-in are resolved and a huge computation time is saved. [Copyright &y& Elsevier]

This paper describes Olex, a novel method for the automatic induction of rule-based text classifiers. Olex supports a hypothesis language of the form "if T1 or … or Tn occurs in document d, and none of Tn+1,…Tn+m occurs in d, then classify d under category c," where each Ti is a conjunction of terms. The proposed method is simple and elegant. Despite this, the results of a systematic experimentation performed on the REUTERS-21578, the ORSUMED, and the ODP data collections show that Olex provides classifiers that are accurate, compact, and comprehensible. A comparative analysis conducted against some of the most well-known learning algorithms (namely, Naive Bayes, Ripper, C4.5, SVM, and Linear Logistic Regression) demonstrates that it is more than competitive in terms of both predictive accuracy and efficiency. [ABSTRACT FROM AUTHOR]

Matrix decomposition methods represent a data matrix as a product of two factor matrices: one containing basis vectors that represent meaningful concepts in the data and another describing how the observed data can be expressed as combinations of the basis vectors. Decomposition methods have been studied extensively, but many methods return real-valued matrices. Interpreting real-valued factor matrices is hard if the original data is Boolean. In this paper, we describe a matrix decomposition formulation for Boolean data, the Discrete Basis Problem. The problem seeks for a Boolean decomposition of a binary matrix, thus allowing the user to easily interpret the basis vectors. We also describe a variation of the problem, the Discrete Basis Partitioning Problem. We show that both problems are NP-hard. For the Discrete Basis Problem, we give a simple greedy algorithm for solving it; for the Discrete Basis Partitioning Problem, we show how it can be solved using existing methods. We present experimental results for the greedy algorithm and compare it against other well-known methods. Our algorithm gives intuitive basis vectors, but its reconstruction error is usually larger than with the real-valued methods. We discuss the reasons for this behavior. [ABSTRACT FROM AUTHOR]

In medical diagnosis, doctors must often determine what medical tests (e.g., X-ray and blood tests) should be ordered for a patient to minimize the total cost of medical tests and misdiagnosis. In this paper, we design cost-sensitive machine learning algorithms to model this learning and diagnosis process. Medical tests are like attributes in machine learning whose values may be obtained at a cost (attribute cost), and misdiagnoses are like misclassifications which may also incur a cost (misclassification cost). We first propose a lazy decision tree learning algorithm that minimizes the sum of attribute costs and misclassification costs. Then, we design several novel "test strategies" that can request to obtain values of unknown attributes at a cost (similar to doctors' ordering of medical tests at a cost) in order to minimize the total cost for test examples (new patients). These test strategies correspond to different situations in real-world diagnoses. We empirically evaluate these test strategies, and show that they are effective and outperform previous methods. Our results can be readily applied to real-world diagnosis tasks. A case study on heart disease is given throughout the paper. [ABSTRACT FROM AUTHOR]

The introduction of next generation sequencing in ecological studies has created a major revolution in microbial and fungal ecology. Direct sequencing of hypervariable regions from ribosomal RNA genes can provide rapid and inexpensive analysis for ecological communities. In order to get deep understanding from these rRNA fragments, the Ribosomal Database Project developed the 'RDP Classifier' utilizing 8-mer nucleotide frequencies with Bayesian theorem to obtain taxonomy affiliation. The classifier is computationally efficient and works well with massive short sequences. However, the binary model employed in the RDP classifier does not consider the repetitive 8-mers in each reference sequence. Previous studies have pointed out that multinomial model usually results a better performance than binary model. In this study, we present the naïve Bayesian classifiers with multinomial models that take repetitive 8-mers into account for classifying microbial 16S and fungal 28S rRNA sequences. The results obtained from the multinomial approach were compared with those obtained from the binomial RDP classifier by 250-bp, 400-bp, 800-bp, and full-length reads to demonstrate that the multinomial approach can generally achieve a higher prediction accuracy in most hypervariable regions. [ABSTRACT FROM PUBLISHER]

In biomedical data, the imbalanced data problem occurs frequently and causes poor prediction performance for minority classes. It is because the trained classifiers are mostly derived from the majority class. In this paper, we describe an ensemble learning method combined with active example selection to resolve the imbalanced data problem. Our method consists of three key components: 1) an active example selection algorithm to choose informative examples for training the classifier, 2) an ensemble learning method to combine variations of classifiers derived by active example selection, and 3) an incremental learning scheme to speed up the iterative training procedure for active example selection. We evaluate the method on six real-world imbalanced data sets in biomedical domains, showing that the proposed method outperforms both the random under sampling and the ensemble with under sampling methods. Compared to other approaches to solving the imbalanced data problem, our method excels by 0.03-0.15 points in AUC measure. [ABSTRACT FROM PUBLISHER]

While genome sequencing projects have generated tremendous amounts of protein sequence data for a vast number of genomes, substantial portions of most genomes are still unannotated. Despite the success of experimental methods for identifying protein functions, they are often lab intensive and time consuming. Thus, it is only practical to use in silico methods for the genome-wide functional annotations. In this paper, we propose new features extracted from protein sequence only and machine learning-based methods for computational function prediction. These features are derived from a position-specific scoring matrix, which has shown great potential in other bininformatics problems. We evaluate these features using four different classifiers and yeast protein data. Our experimental results show that features derived from the position-specific scoring matrix are appropriate for automatic function annotation. [ABSTRACT FROM PUBLISHER]

The proposed system combines unsupervised clustering, social network analysis, semisupervised feature induction, and supervised classification to model user priorities among incoming email messages. [ABSTRACT FROM PUBLISHER]