Mining Classification Rules from Datasets with Large Number of Many-Valued Attributes

Abstract

Abstract. Decision tree induction algorithms scale well to large datasets for their univariate and divide-and-conquer approach. However, they may fail in discovering effective knowledge when the input dataset consists of a large number of uncorrelated many-valued attributes. In this paper we present an algorithm, Noah, that tackles this problem by applying a multivariate search. Performing a multivariate search leads to a much larger consumption of computation time and memory, this may be prohibitive for large datasets. We remedy this problem by exploiting effective pruning strategies and efficient data structures. We applied our algorithm to a real marketing application of cross-selling. Experimental results revealed that the application database was too complex for C4.5 as it failed to discover any useful knowledge. The application database was also too large for various well known rule discovery algorithms which were not able to complete their task. The pruning techniques used in Noah are general in nature and can be used in other mining systems.

Modeling Customer Lifetime Value

“…In their simulation, the overall mean prediction rate of the logit was 72.7%, whereas the hit rate for SVM was 85.9%. Similarly, Giuffrida, Chu, and Hanssens (2000) reported that a multivariate decision tree induction algorithm outperformed a logit model in identifying the best customer targets for cross-selling purposes.…”

Privacy-Preserving Data Publishing Based on Utility Specification

“…We used all 45,222 tuples in the Adult dataset and 12,960 tuples in the Nursery dataset. We considered decision tree [26], Naive Bayes classifier [27], [28], and classification rules [29], [30] as knowledge models. We used the classification accuracy to measure the quality of decision trees and Naive Bayes classifiers, and the number of preserved classification rules to measure classification rules.…”

Retailer promotion planning: Improving forecast accuracy and interpretability

“…Algorithm application to forecasting and direct marketing has been discussed in management and computer science literature (Cooper & Giuffrida, 2000;Giuffrida, Chu, & Hanssens, 2000). The authors and their affiliates conducted extensive benchmark testing of the KDS/Noah algorithm in application to promotion forecasting and demonstrated its superiority to multiple commercially available data mining solutions including SAS EnterpriseMiner, SOMine, CN2, Ripper, Apriory, CBA, and others (Krycha, 1999).…”