A hierarchical multiple kernel support vector machine for customer churn prediction using longitudinal behavioral data

The availability of abundant data posts a challenge to integrate static customer data and longitudinal behavioral data to improve performance in customer churn prediction. Usually, longitudinal behavioral data are transformed into static data before being included in a prediction model. In this study, a framework with ensemble techniques is presented for customer churn prediction directly using longitudinal behavioral data. A novel approach called the hierarchical multiple kernel support vector machine (H-MK-SVM) is formulated. A three phase training algorithm for the H-MK-SVM is developed, implemented and tested. The H-MK-SVM constructs a classification function by estimating the coefficients of both static and longitudinal behavioral variables in the training process without transformation of the longitudinal behavioral data. The training process of the H-MK-SVM is also a feature selection and time subsequence selection process because the sparse non-zero coefficients correspond to the variables selected. Computational experiments using three real-world databases were conducted. Computational results using multiple criteria measuring performance show that the H-MK-SVM directly using longitudinal behavioral data performs better than currently available classifiers.     

Cost,revenue and profit efficiency measurement in DEA: A directional distance function approach

Estimation of efficiency of firms in a non-competitive market characterized by heterogeneous inputs and outputs along with their varying prices is questionable when factor-based technology sets are used in data envelopment analysis (DEA). In this scenario, a value-based technology becomes an appropriate reference technology against which efficiency can be assessed. In this contribution, the value-based models of Tone (2002) are extended in a directional DEA set up to develop new directional cost- and revenue-based measures of efficiency, which are then decomposed into their respective directional value-based technical and allocative efficiencies. These new directional value-based measures are more general, and include the existing value-based measures as special cases. These measures satisfy several desirable properties of an ideal efficiency measure. These new measures are advantageous over the existing ones in terms of (1) their ability to satisfy the most important property of translation invariance; (2) choices over the use of suitable direction vectors in handling negative data; and (3) flexibility in providing the decision makers with the option of specifying preferable direction vectors to incorporate their preferences. Finally, under the condition of no prior unit price information, a directional value-based measure of profit inefficiency is developed for firms whose underlying objectives are profit maximization. For an illustrative empirical application, our new measures are applied to a real-life data set of 50 US banks to draw inferences about the production correspondence of banking industry.     

Hybridizing principles of the Electre method with case-based reasoning for data mining: Electre-CBR-I and Electre-CBR-II

Electre is an important outranking method developed in the area of decision-aiding. Data mining is a vital developing technique that receives contributions from lots of disciplines such as databases, machine learning, information retrieval, statistics, and so on. Techniques in outranking approaches, e.g. Electre, could also contribute to the development of data mining. In this research, we address the following two issues: a) why and how to combine Electre with case-based reasoning (CBR) to generate corresponding hybrid models by extending the fundamental principles of Electre into CBR; b) the effect on predictive performance by employing evidence vetoing the assertion on the base of evidence supporting the assertion. The similarity measure of CBR is implemented by revising and fulfilling three basic ideas of Electre, i.e. assertion that two cases are indifferent, evidence supporting the assertion, and evidence vetoing the assertion. Two corresponding CBR models are constructed by combining principles of the Electre decision-aiding method with CBR. The first one, named Electre-CBR-I, derives from evidence supporting the assertion. The other one, named Electre-CBR-II, derives from both evidence supporting and evidence vetoing the assertion. Leave-one-out cross-validation and hold-out method are integrated to form 30-times hold-out method. In financial distress mining, data was collected from Shanghai and Shenzhen Stock Exchanges, ANOVA was employed to select features that are significantly different between companies in distress and health, 30-times hold-out method was used to assess predictive performance, and grid-search technique was utilized to search optimal parameters. Original data distributions were kept in the experiment. Empirical results of long-term financial distress prediction with 30 initial financial ratios and 135 initial pairs of samples indicate that Electre-CBR-I outperforms Electre-CBR-II and other comparative CBR models, and Electre-CBR-II outperforms the other comparative CBR models.     

New approach for the assignment of new European agricultural subsidies using scores from data envelopment analysis: Application to olive-growing farms in Andalusia (Spain)

More than 60% of olive-growing farms in Andalusia (Spain) would have negative returns without European agricultural subsidies. Agenda 2000 criteria imply that agricultural subsidies currently play the role of enhancing the production quality and the environmental and social values of agriculture. Although the necessity of the modulation of subsidies is stated, the EU regulations do not state which objective criteria should be used or how they should be measured, despite the fact that regulations demand objectivity in this measurement.