The composition of simple local models for approximating complex nonlinear mappings is a common practice in recent modeling and control literature. This paper presents a comparative analysis of two different local approaches: the neuro-fuzzy inference system and the lazy learning approach. Neuro-fuzzy is a hybrid representation which combines the linguistic description typical of fuzzy inference systems, with learning procedures inspired by neural networks. Lazy learning is a memory-based technique that uses a query-based approach to select the best local model configuration by assessing and comparing different alternatives in cross-validation. In this paper, the two approaches are compared both as learning algorithms, and as identification modules of an adaptive control system. We show that lazy learning is able to provide better modeling accuracy and higher control performance at the cost of a reduced readability of the resulting approximator. Illustrative examples of identi cation and control of a nonlinear system starting from simulated data are given.

Abstract. This paper proposes a methodology named OP-ELM, based on a recent development –the Extreme Learning Machine – decreasing drastically the training speed of networks. Variable selection is beforehand performed on the original dataset for proper results by OP-ELM: the network is first created using Extreme Learning Process, selection of the most relevant nodes is performed using Least Angle Regression (LARS) ranking of the nodes and a Leave-One-Out estimation of the performances. Results are globally equivalent to LSSVM ones with reduced computational time. 1

Abstract not found

Abstract — Bankruptcy prediction have been widely studied as a binary classification problem using financial ratios methodologies. When calculating the ratios, it is common to confront missing data problem. Thus, this paper proposes a classification method Ensemble Nearest Neighbors (ENN) to solve it. ENN uses different nearest neighbors as ensemble classifiers, then make a linear combination of them. Instead of choosing k in original k-Nearest Neighbors algorithm, ENN provides weights to each classifier which makes the method more robust. Moreover, using a adapted distance metric, ENN can be used directly for incomplete data. In a word, ENN is a robust and a comparatively simple model while providing good performance with missing data. In the experiment section, four financial datasets which are publicly available are used to prove this conclusion.

Abstract. This paper presents a working combination of input selection strategy and a fast approximator for time series prediction. The input selection is performed using Tabu Search with the Delta Test. The approximation methodology is called Optimally-Pruned k-Nearest Neighbors (OP-KNN), which has been recently developed for fast and accurate regression and classification tasks. In this paper we demonstrate the accuracy of the OP-KNN with the Tabu Search using the ESTSP 2008 Competition datasets.