Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Cost-sensitive meta-learning framework

Cost-sensitive meta-learning framework This 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. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Modelling in Management Emerald Publishing

Cost-sensitive meta-learning framework

Download PDF
21 pages

Loading next page...
 
/lp/emerald-publishing/cost-sensitive-meta-learning-framework-j4xny0nZn0
Publisher
Emerald Publishing
Copyright
© Emerald Publishing Limited
ISSN
1746-5664
eISSN
1746-5664
DOI
10.1108/jm2-03-2021-0065
Publisher site
See Article on Publisher Site

Abstract

This 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.

Journal

Journal of Modelling in Management – Emerald Publishing

Published: Aug 22, 2022

Keywords: Artificial intelligence; Modeling; Decision analysis; Data mining; Cost analysis; Meta-learning; Cost-sensitive algorithms; Machine learning algorithms

References

  • Speeding up algorithm selection using average ranking and active testing by introducing runtime
    Abdulrahman, S.M.; Brazdil, P.; van Rijn, J.N.; Vanschoren, J.
  • Generalizing from case studies: a case study
    Aha, D.W.
  • MFE: towards reproducible meta-feature extraction
    Alcobaça, E.; Siqueira, F.; Rivolli, A.; Garcia, L.P.; Oliva, J.T.; de Carvalho, A.C.
  • On learning algorithm selection for classification
    Ali, S.; Smith, K.A.
  • UCI machine learning repository
    Bache, K.; Lichman, M.
  • Using Meta-learning to recommend process discovery methods
    Barbon, S.; Ceravolo, P.; Damiani, E.; Tavares, G.M.
  • Estimating the predictive accuracy of a classifier
    Bensusan, H.; Kalousis, A.
  • Evaluation of machine-learning algorithm ranking advisors
    Berrer, H.; Paterson, I.; Keller, J.
  • Ranking learning algorithms: using IBL and Meta-learning on accuracy and time results
    Brazdil, P.; Soares, C.; Da Costa, J.P.
  • Revisiting Meta-learning as supervised learning
    Chao, W.L.; Ye, H.J.; Zhan, D.C.; Campbell, M.; Weinberger, K.Q.
  • Metalearning and recommender systems: a literature review and empirical study on the algorithm selection problem for collaborative filtering
    Cunha, T.; Soares, C.; de Carvalho, A.C.
  • Cost-sensitive business failure prediction when misclassification costs are uncertain: a heterogeneous ensemble selection approach
    De Bock, K.W.; Coussement, K.; Lessmann, S.
  • Correlation-based oversampling aided cost sensitive ensemble learning technique for treatment of class imbalance
    Devi, D.; Biswas, S.K.; Purkayastha, B.
  • Metacost: a general method for making classifiers cost-sensitive
    Domingos, P.
  • The foundations of cost-sensitive learning
    Elkan, C.
  • Using cost-sensitive learning and feature selection algorithms to improve the performance of imbalanced classification
    Feng, F.; Li, K.C.; Shen, J.; Zhou, Q.; Yang, X.
  • Cost-sensitive learning
    Fernández, A.; García, S.; Galar, M.; Prati, R.C.; Krawczyk, B.; Herrera, F.
  • Information, prediction, and query by committee
    Freund, Y.; Seung, H.S.; Shamir, E.; Tishby, N.
  • Integrating TANBN with cost sensitive classification algorithm for imbalanced data in medical diagnosis
    Gan, D.; Shen, J.; An, B.; Xu, M.; Liu, N.
  • Noise detection in the Meta-learning level
    Garcia, L.P.; de Carvalho, A.C.; Lorena, A.C.
  • The WEKA data mining software: an update
    Hall, M.; Frank, E.; Holmes, G.; Pfahringer, B.; Reutemann, P.; Witten, I.H.
  • Meta-learning in neural networks: a survey
    Hospedales, T.; Antoniou, A.; Micaelli, P.; Storkey, A.
  • Machine learning models and cost-sensitive decision trees for bond rating prediction
    Jabeur, S.B.; Sadaaoui, A.; Sghaier, A.; Aloui, R.
  • Noemon: design, implementation and performance results of an intelligent assistant for classifier selection
    Kalousis, A.; Theoharis, T.
  • A systematic review on imbalanced data challenges in machine learning: applications and solutions
    Kaur, H.; Pannu, H.S.; Malhi, A.K.
  • A literature survey and empirical study of Meta-learning for classifier selection
    Khan, I.; Zhang, X.; Rehman, M.; Ali, R.
  • Statlog: comparison of classification algorithms on large real-world problems
    King, D.; Feng, C.; Sutherland, A.
  • Handling imbalanced datasets: a review
    Kotsiantis, S.; Kanellopoulos, D.; Pintelas, P.
  • Predicting duration of traffic accidents based on cost-sensitive Bayesian network and weighted K-nearest neighbor
    Kuang, L.; Yan, H.; Zhu, Y.; Tu, S.; Fan, X.
  • Instance categorization by support vector machines to adjust weights in AdaBoost for imbalanced data classification
    Lee, W.; Jun, C.H.; Lee, J.S.
  • Oversampling techniques for bankruptcy prediction: novel features from a transaction dataset
    Le, T.; Lee, M.Y.; Park, J.R.; Baik, S.W.
  • Modified decision tree learning for Cost-Sensitive credit card fraud detection model
    Lenka, S.R.; Barik, R.K.; Patra, S.S.; Singh, V.P.
  • On the use of meta-learning for instance selection: an architecture and an experimental study
    Leyva, E.; Caises, Y.; González, A.; Pérez, R.
  • A survey of cost-sensitive decision tree induction algorithms
    Lomax, S.; Vadera, S.
  • Does cost-sensitive learning beat sampling for classifying rare classes?
    McCarthy, K.; Zabar, B.; Weiss, G.
  • Fault diagnosis method of check valve based on multikernel cost-sensitive extreme learning machine
    Ma, J.; Wu, J.; Wang, X.
  • Cost-sensitive BERT for generalisable sentence classification with imbalanced data
    Madabushi, H.T.; Kochkina, E.; Castelle, M.
  • Cost-sensitive weighting and imbalance-reversed bagging for streaming imbalanced and concept drifting in electricity pricing classification
    Ng, W.W.; Zhang, J.; Lai, C.S.; Pedrycz, W.; Lai, L.L.; Wang, X.
  • Generating better decision trees
    Norton, S.W.
  • A comprehensive overview and survey of recent advances in meta-Learning
    Peng, H.
  • Cost-sensitive learning for profit-driven credit scoring
    Petrides, G.; Moldovan, D.; Coenen, L.; Guns, T.; Verbeke, W.
  • Prediction of diabetes using cost sensitive learning and oversampling techniques on Bangladeshi and Indian female patients
    Pranto, B.; Mehnaz, S.M.; Momen, S.; Huq, S.M.
  • Selecting machine learning algorithms using the ranking meta-learning approach
    Prudancio, R.; De Souto, M.C.P.; Ludermir, T.B.
  • Automatic classifier selection for non-experts
    Reif, M.; Shafait, F.; Goldstein, M.; Breuel, T.; Dengel, A.
  • Empirical learning as a function of concept character
    Rendell, L.; Cho, H.
  • Improving the design of induction methods by analyzing algorithm functionality and data-based concept complexity
    Rendell, L.; Ragavan, H.
  • The algorithm selection problem
    Rice, J.R.
  • MetaStream: a meta-learning based method for periodic algorithm selection in time-changing data
    Rossi, A.L.D.; de Leon Ferreira, A.C.P.; Soares, C.; De Souza, B.F.
  • Improved boosting algorithms using confidence-rated predictions
    Schapire, R.E.; Singer, Y.
  • A novel meta learning framework for feature selection using data synthesis and fuzzy similarity
    Shen, Z.; Chen, X.; Garibaldi, J.M.
  • Feature selection in meta learning framework
    Shilbayeh, S.; Vadera, S.
  • Meta decision trees for explainable recommendation systems
    Shulman, E.; Wolf, L.
  • Class imbalance and cost-sensitive decision trees: a unified survey based on a core similarity
    Siers, M.J.; Islam, M.Z.
  • Cross-disciplinary perspectives on meta-learning for algorithm selection
    Smith-Miles, K.A.
  • Automatic recommendation of classification algorithms based on data set characteristics
    Song, Q.; Wang, G.; Wang, C.
  • Classification of imbalanced data: a review
    Sun, Y.; Wong, A.K.C.; Kamel, M.S.
  • Cost-sensitive concept learning of sensor use in approach and recognition
    Tan, M.; Schlimmer, J.
  • Boosting methods for multi-class imbalanced data classification: an experimental review
    Tanha, J.; Abdi, Y.; Samadi, N.; Razzaghi, N.; Asadpour, M.
  • Self-adaptive cost weights-based support vector machine cost-sensitive ensemble for imbalanced data classification
    Tao, X.; Li, Q.; Guo, W.; Ren, C.; Li, C.; Liu, R.; Zou, J.
  • Data imbalance in classification: Experimental evaluation
    Thabtah, F.; Hammoud, S.; Kamalov, F.; Gonsalves, A.
  • Inducing cost-sensitive trees via instance weighting
    Ting, K.M.
  • A study of algorithm selection in data mining using Meta-Learning
    Tripathy, M.; Panda, A.
  • Cost-sensitive classification: empirical evaluation of a hybrid genetic decision tree induction algorithm
    Turney, P.
  • Types of cost in inductive concept learning
    Turney, P.
  • Sampling approaches for imbalanced data classification problem in machine learning
    Tyagi, S.; Mittal, S.
  • A feature subset selection algorithm automatic recommendation method
    Wang, G.; Song, Q.; Sun, H.; Zhang, X.; Xu, B.; Zhou, Y.
  • No free lunch theorems for optimization. Evolutionary computation
    Wolpert, D.H.; Macready, W.G.
  • Application of alternating decision tree with AdaBoost and bagging ensembles for landslide susceptibility mapping
    Wu, Y.; Ke, Y.; Chen, Z.; Liang, S.; Zhao, H.; Hong, H.
  • Cost-sensitive boosted tree for loan evaluation in peer-to-peer lending
    Xia, Y.; Liu, C.; Liu, N.
  • Cost-sensitive learning by cost-proportionate example weighting
    Zadrozny, B.; Langford, J.; Abe, N.
  • Example-dependent cost-sensitive adaptive boosting
    Zelenkov, Y.
  • Cost-sensitive KNN classification
    Zhang, S.
  • Imbalanced classification of mental workload using a cost-sensitive majority weighted minority oversampling strategy
    Zhang, J.; Cui, X.; Li, J.; Wang, R.
  • An instance-based learning recommendation algorithm of imbalance handling methods
    Zhang, X.; Li, R.; Zhang, B.; Yang, Y.; Guo, J.; Ji, X.
  • A cost sensitive decision tree algorithm based on weighted class distribution with batch deleting attribute mechanism
    Zhao, H.; Li, X.
  • A model of inductive bias learning
    Baxter, J.
  • Overfitting and undercomputing in machine learning
    Dietterich, T.
  • A hybrid approach using oversampling technique and cost-sensitive learning for bankruptcy prediction
    Le, T.; Vo, M.T.; Vo, B.; Lee, M.Y.; Baik, S.W.
  • The use of background knowledge in decision tree induction
    Núnez, M.