Discounted least squares-improved circular back-propogation neural networks with applications in time series prediction

Summary As a generalization of the multi-layer perceptron (MLP), the circular back-propagation neural network (CBP) possesses better adaptability. An improved version of the CBP (the ICBP) is presented in this paper. Despite having less adjustable weights, the ICBP has better adaptability than the CBP, which quite equals the famous Occams razor principle for model selection. In its application to time series, considering both structural changes and correlations of time series itself, we introduce the principle of the discounted least squares (DLS) in CBP and ICBP, respectively, and investigate their predicting capacity further. Introduction of DLS improves the predicting performance of both on a benchmark time series data set. Finally, the comparison of experimental results shows that ICBP with DLS (DLS-ICBP) has better predicting performance than DLS-CBP.Supported by Natural Science (grant No.BK2002092) and QingLan project foundations of Jiangsu province and Returnee foundation of China.     

Mining Customer Value: From Association Rules to Direct Marketing

Direct marketing is a modern business activity with an aim to maximize the profit generated from marketing to a selected group of customers. A key to direct marketing is to select a subset of customers so as to maximize the profit return while minimizing the cost. Achieving this goal is difficult due to the extremely imbalanced data and the inverse correlation between the probability that a customer responds and the dollar amount generated by a response. We present a solution to this problem based on a creative use of association rules. Association rule mining searches for all rules above an interestingness threshold, as opposed to some rules in a heuristic-based search. Promising association rules are then selected based on the observed value of the customers they summarize. Selected association rules are used to build a model for predicting the value of a future customer. On the challenging KDD-CUP-98 dataset, this approach generates 41% more profit than the KDD-CUP winner and 35% more profit than the best result published thereafter, with 57.7% recall on responders and 78.0% recall on non-responders. The average profit per mail is 3.3 times that of the KDD-CUP winner.     

Theory and experiment on the measurement of approach-avoidance conflict.

To investigate the effect of approach-avoidance conflict upon GSR and the content and reaction time in a word association task, parachutists were tested on the day of a scheduled jump and either 2 weeks before or after the jump. Gradients of anxiety were higher on the day of the jump for all Ss, as measured by the dependent variables, as compared to the pre- and postjump periods. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Continual association as a function of level of creativity and type of verbal stimulus.

Continual word association was studied as a function of Remote Associates Test (RAT) performance, form class, associative hierarchy, and Thorndike-Lorge word frequency. Ss were selected as high creative (HC), low creative (LC), and medium creative (MC) on the basis of their RAT scores. It was found that HC Ss give the greatest number of associations and maintain a relatively higher speed of association throughout a 2-min. period. More responses were elicited by nouns than adjectives, flat hierarchy words than steep, and high frequency words than low. The S and stimulus variable did not interact. Relevance to an associative theory of creative thinking is discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The predictive power of ridge regression: Some quasi-simulation results: Correction to Cattin.

Reports an error in the original article by Philippe Cattin (Journal of Applied Psychology, 1981, Vol. 66, No. 3, pp. 282-290). The fifth sentence in the first paragraph on page 284 contains an error. The sentence should read: "For intermediate values of k?, the ridge regression weights are 'weighted sums' of the OLS regression weights [not models] and of the zero-order sample correlations and tend to decrease in absolute value as k? increases." (The following abstract of this article originally appeared in record 1981-27117-001.) Reviews research indicating that ridge regression tends to improve the mean square error of prediction obtained with ordinary least squares (OLS) regression in a wide range of conditions. MMPI profiles of 861 psychiatric patients and diagnoses from 29 psychologists, as collected by P. E. Meehl (see record 1960-04396-001), were used to examine the gains in cross-validated multiple correlation obtained with ridge regression compared with OLS and equal weights as a function of sample size and ridge constant. A simple formula for estimating the ridge constant was also evaluated. Results that are related to recent developments concerning the use of Bayesian regression procedures show that ridge regression improves both the mean square error of prediction and the cross-validated multiple correlation obtained with OLS when the ratio sample size to number of predictor variables is relatively small. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Multivariable system structure and parameter identification using the correlation method

This paper describes an algorithm for the structure determination and parameter identification of linear discrete-time multivariable systems from input-output measurements. The algorithm starts by determining the structure parameters of a certain canonical state space representation from an estimate of the correlation functions of the output sequence. Then the parameters of the A matrix are estimated from the estimated correlation functions using the recursive least squares method. Finally a normalized stochastic approximation algorithm is used for the estimation of the parameters of the B matrix from input-output measurements.     

Mining churning behaviors and developing retention strategies based on a partial least squares (PLS) model

In a very competitive mobile telecommunication business environment, marketing managers need a business intelligence model that allows them to maintain an optimal (at least a near optimal) level of churners very effectively and efficiently while minimizing the costs throughout their marketing programs. As a first step toward optimal churn management program for marketing managers, this paper focuses on building an accurate and concise predictive model for the purpose of churn prediction utilizing a partial least squares (PLS)-based methodology on highly correlated data sets among variables. A preliminary experiment demonstrates that the presented model provides more accurate performance than traditional prediction models and identifies key variables to better understand churning behaviors. Further, a set of simple churn marketing programs—device management, overage management, and complaint management strategies—is presented and discussed.     

Enhancing the semantics of UML association redefinition

Association redefinition is a UML construct that permits us to define an association end more specifically in a particular context. Concretely, it allows specifying some additional participation and cardinality constraints on the association. Association refinements, which have been studied and used by many authors in conceptual modelling languages prior to UML, are closely related to association redefinitions. They also permit to refine the ends of an association adding participation and cardinality constraints. In this paper, we analyze and compare the semantics of both concepts and propose to extend the semantics of association redefinitions in UML to cover all the constraints that may be expressed by association refinements in other conceptual modelling languages. Additionally, we present how to integrate previous results on validation of association refinements to UML and how to generate code for a relational technology platform. Finally, we provide a prototype tool to verify the feasibility of the approach.     

Reliable representations for association rules

Association rule mining has contributed to many advances in the area of knowledge discovery. However, the quality of the discovered association rules is a big concern and has drawn more and more attention recently. One problem with the quality of the discovered association rules is the huge size of the extracted rule set. Often for a dataset, a huge number of rules can be extracted, but many of them can be redundant to other rules and thus useless in practice. Mining non-redundant rules is a promising approach to solve this problem. In this paper, we first propose a definition for redundancy, then propose a concise representation, called a Reliable basis, for representing non-redundant association rules. The Reliable basis contains a set of non-redundant rules which are derived using frequent closed itemsets and their generators instead of using frequent itemsets that are usually used by traditional association rule mining approaches. An important contribution of this paper is that we propose to use the certainty factor as the criterion to measure the strength of the discovered association rules. Using this criterion, we can ensure the elimination of as many redundant rules as possible without reducing the inference capacity of the remaining extracted non-redundant rules. We prove that the redundancy elimination, based on the proposed Reliable basis, does not reduce the strength of belief in the extracted rules. We also prove that all association rules, their supports and confidences, can be retrieved from the Reliable basis without accessing the dataset. Therefore the Reliable basis is a lossless representation of association rules. Experimental results show that the proposed Reliable basis can significantly reduce the number of extracted rules. We also conduct experiments on the application of association rules to the area of product recommendation. The experimental results show that the non-redundant association rules extracted using the proposed method retain the same inference capacity as the entire rule set. This result indicates that using non-redundant rules only is sufficient to solve real problems needless using the entire rule set.