Cost-sensitive Classification Methods Research Articles

Cost-sensitive meta-learning framework

PurposeThis paper aims to describe the use of a meta-learning framework for recommending cost-sensitive classification methods with the aim of answering an important question that arises in machine learning, namely, “Among all the available classification algorithms, and in considering a specific type of data and cost, which is the best algorithm for my problem?”Design/methodology/approachThis paper describes the use of a meta-learning framework for recommending cost-sensitive classification methods for the aim of answering an important question that arises in machine learning, namely, “Among all the available classification algorithms, and in considering a specific type of data and cost, which is the best algorithm for my problem?” The framework is based on the idea of applying machine learning techniques to discover knowledge about the performance of different machine learning algorithms. It includes components that repeatedly apply different classification methods on data sets and measures their performance. The characteristics of the data sets, combined with the algorithms and the performance provide the training examples. A decision tree algorithm is applied to the training examples to induce the knowledge, which can then be used to recommend algorithms for new data sets. The paper makes a contribution to both meta-learning and cost-sensitive machine learning approaches. Those both fields are not new, however, building a recommender that recommends the optimal case-sensitive approach for a given data problem is the contribution. The proposed solution is implemented in WEKA and evaluated by applying it on different data sets and comparing the results with existing studies available in the literature. The results show that a developed meta-learning solution produces better results than METAL, a well-known meta-learning system. The developed solution takes the misclassification cost into consideration during the learning process, which is not available in the compared project.FindingsThe proposed solution is implemented in WEKA and evaluated by applying it to different data sets and comparing the results with existing studies available in the literature. The results show that a developed meta-learning solution produces better results than METAL, a well-known meta-learning system.Originality/valueThe paper presents a major piece of new information in writing for the first time. Meta-learning work has been done before but this paper presents a new meta-learning framework that is costs sensitive.

Multi-label thresholding for cost-sensitive classification

Multi-label classification associates each instance with a set of labels which reflects the nature of a wide range of real-world applications. However, existing approaches assume that all labels have the same misclassification cost, whereas in real-world problems different types of misclassification errors have different costs, which are generally unknown in the training context or might change from one context to another. Thus, there is a demand for cost-sensitive classification methods that minimise the average misclassification cost rather than error rates or counts. In this paper, we adopt a simple yet general method, called thresholding, which applies to most classification algorithms to adapt them to cost-sensitive multi-label classification. This paper investigates current threshold choice approaches for multi-label classification. It explores the choice of single and multiple thresholds and extends some of the current techniques to support multi-label problems. Moreover, it proposes cost curves and scatter diagrams for performance evaluation in the multi-label setting. Experimental evaluation on 13 multi-label datasets demonstrates that there is no significant loss by adjusting a global threshold rather than a per-label threshold considering different misclassification costs across labels. Although tuning multiple thresholds is the obvious solution, the global threshold can also be valid.

Hybrid Classifier Ensemble for Imbalanced Data.

The class imbalance problem has become a leading challenge. Although conventional imbalance learning methods are proposed to tackle this problem, they have some limitations: 1) undersampling methods suffer from losing important information and 2) cost-sensitive methods are sensitive to outliers and noise. To address these issues, we propose a hybrid optimal ensemble classifier framework that combines density-based undersampling and cost-effective methods through exploring state-of-the-art solutions using multi-objective optimization algorithm. Specifically, we first develop a density-based undersampling method to select informative samples from the original training data with probability-based data transformation, which enables to obtain multiple subsets following a balanced distribution across classes. Second, we exploit the cost-sensitive classification method to address the incompleteness of information problem via modifying weights of misclassified minority samples rather than the majority ones. Finally, we introduce a multi-objective optimization procedure and utilize connections between samples to self-modify the classification result using an ensemble classifier framework. Extensive comparative experiments conducted on real-world data sets demonstrate that our method outperforms the majority of imbalance and ensemble classification approaches.

CBR-PSO: cost-based rough particle swarm optimization approach for high-dimensional imbalanced problems

Datasets, which have a considerably larger number of attributes compared to samples, face a serious classification challenge. This issue becomes even harder when such high-dimensional datasets are also imbalanced. Recently, such datasets have attracted the interest of both industry and academia and thereby have become a very attractive research area. In this paper, a new cost-sensitive classification method, the CBR-PSO, is presented for such high-dimensional datasets with different imbalance ratios and number of classes. The CBR-PSO is based on particle swarm optimization and rough set theory. The robustness of the algorithm is based on the simultaneously applying attribute reduction and classification; in addition, these two stages are also sensitive to misclassification cost. Algorithm efficiency is examined in publicly available datasets and compared to well-known attribute reduction and cost-sensitive classification algorithms. The statistical analysis and experiments showed that the CBR-PSO can be better than or comparable to the other algorithms, in terms of MAUC values.

An Efficient Misclassification Cost Concerned Imbalanced Class Handling to Improve the Financial Fraud Detection Scheme

Objective: Financial fraud detection is the most concerned task in many of the real world application where the number of fraudulent transaction is occurring. The efficient handling of the imbalanced financial data set would lead to the accurate and improved prediction of the more fraudulent behaviour present in the financial system. Method: An Optimal Ensemble Classification using PSO (OEC-PSO) is introduced. In our previous research work, Forecasting of Automatic Relevance Determination and classification (FARD-RIPPER) approach has been suggested which would classify the relevant subset of features that are more relevant to each other using ripper classification approach. However single base classifier used for classification might lead to the high misclassification rate which is resolved in the proposed research work. makes use of four different classifiers namely, ADTree, Cart, Prism and Ripper for Assembling to obtain accurate classification result. Particle swarm optimization approach is introduced in this work for selecting the most optimal ensemble classifier which can produce the accurate classification result. The fitness value is updated in the every iteration by randomly changing the combination of classifiers and the subset features from training data partitions. Final Ensembling result is obtained by fusing the different classifiers results using weighted average scheme. The overall flow of this work provides flexible way for the users to detect the fraudulent behaviours present in the financial transaction system in terms of malicious patterns. Findings: The experimental tests were conducted in the Matlab simulation environment in terms of various performance metrics to prove the improvement of the proposed approach. The various performance metrics that are considered in this work are accuracy, precision, misclassification cost and the computational complexity. The experimental tests were conducted between the proposed methodology named OEC-PSO and the existing work called the Cost Sensitive Classification Method (CSCM). The results obtained were proved that the proposed methodology can work well with the improved accuracy, misclassification cost and the precision rate. Application/Improvements: The finding of this work demonstrates that the proposed methodology provides better result than the existing work in terms of improved accuracy and the computational procedure. It can be applied in the various online transaction oriented application such as banking, insurance sector and so on.

A cost-sensitive classification algorithm: BEE-Miner

Classification is a data mining technique which is utilized to predict the future by using available data and aims to discover hidden relationships between variables and classes. Since the cost component is crucial in most real life classification problems and most traditional classification methods work for the purpose of correct classification, developing cost-sensitive classifiers which minimize the total misclassification cost remains a subject of much interest. The purpose of this study is to present an effective solution method that configurates and evaluates learning systems from previous experiences, thus aiming to obtain decisions and predictions. Since most real life problems are cost-sensitive and developing effective direct methods for cost-sensitive multi-class classification is still an attractive area, a cost-sensitive classification method, the BEE-Miner algorithm, is proposed by utilizing the recently developed Bees Algorithm (BA). The main advantages of BEE-Miner are its capability to handle both binary and multi-class problems and to incorporate misclassification cost into the algorithm via generating neighbor solutions and evaluating the quality of the solutions. An extensive computational study is also performed on cost-insensitive and cost-sensitive versions of the proposed BEE-Miner algorithm and effective results on different types of problems are obtained with high test accuracy and low misclassification cost.

Software defect prediction using a cost sensitive decision forest and voting, and a potential solution to the class imbalance problem

Software development projects inevitably accumulate defects throughout the development process. Due to the high cost that defects can incur, careful consideration is crucial when predicting which sections of code are likely to contain defects. Classification algorithms used in machine learning can be used to create classifiers which can be used to predict defects. While traditional classification algorithms optimize for accuracy, cost-sensitive classification methods attempt to make predictions which incur the lowest classification cost. In this paper we propose a cost-sensitive classification technique called CSForest which is an ensemble of decision trees. We also propose a cost-sensitive voting technique called CSVoting in order to take advantage of the set of decision trees in minimizing the classification cost. We then investigate a potential solution to class imbalance within our decision forest algorithm. We empirically evaluate the proposed techniques comparing them with six (6) classifier algorithms on six (6) publicly available clean datasets that are commonly used in the research on software defect prediction. Our initial experimental results indicate a clear superiority of the proposed techniques over the existing ones.