Selecting Machine Learning Algorithms Using the Ranking Meta-Learning Approach Ricardo B. C. Prudˆ encio, Marcilio C. P. de Souto, and Teresa B. Ludermir Center of Informatics, Federal University of Pernambuco, Cidade Universit´ aria - CEP 50732-970 - Recife (PE) - Brazil {rbcp, mcps, tbl}@cin.ufpe.br Abstract. In this work, we present the use of Ranking Meta-Learning approaches to ranking and selecting algorithms for problems of time se- ries forecasting and clustering of gene expression data. Given a problem (forecasting or clustering), the Meta-Learning approach provides a rank- ing of the candidate algorithms, according to the characteristics of the problem’s dataset. The best ranked algorithm can be returned as the selected one. In order to evaluate the Ranking Meta-Learning proposal, prototypes were implemented to rank artificial neural networks models for forecasting financial and economic time series and to rank clustering algorithms in the context of cancer gene expression microarray datasets. The case studies regard experiments to measure the correlation between the suggested rankings of algorithms and the ideal rankings. The results revealed that Meta-Learning was able to suggest more adequate rankings in both domains of application considered. 1 Introduction One of the major challenges in many domains of Computational Intelligence, Machine Learning, Data Analysis and other fields is to investigate the capabili- ties and limitations of the existing algorithms in order to identify when one algo- rithm is more adequate than another to solve particular problems [1]. Traditional approaches to selecting algorithms involve, in general, costly trial-and-error pro- cedures, or require expert knowledge, which is not always easy to acquire [2]. Meta-Learning for algorithm selection arises in this context as an effective solu- tion, capable of automatically predicting algorithm’s performance, thus assisting users in the choice of the most adequate techniques for dealing with the problems at hand [2–6]. In Meta-Learning, each meta-example is related to a learning problem and stores: (1) the features describing the problem, called meta-features; and (2) the performance information about one or more algorithms when applied to the problem. By receiving a set of such meta-examples, another learning system (the meta-learner) is applied to acquire knowledge relating the performance of the candidate algorithms and the descriptive features of the problems. The acquired knowledge can then be used to predict algorithm performance for new prob- lems not seen during the Meta-Learning process and to recommend algorithms.