Zhenjiang Hu, Wei-Ngan Chin, and Masato Takeichi. Calculating a New Data Mining Algorithm for Market Basket Analysis. Lecture Notes in Computer Science, 1753, 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....management performance bottleneck. Sequential levelwise algorithms for finding frequent sets have been well studied [1, 9] although performance increase of two orders of magnitude have been recently reported by clever strategies for representing and generating large candidates from smaller ones [7]. These algorithms are relatively straightforward to parallelize, although a number of different strategies have been used to reduce the extra communication that arises from parallelizing storage manipulation in the sequential case [5] We show that approximate sampling algorithms can be adapted ....

....Thus the computation required for the new algorithm is only slightly greater than that of a levelwise algorithm. Notice that all of the clever data structuring and incremental construction 8 techniques used in sequential levelwise algorithms can be immediately applied to the parallel case [7]. 5 Handling Large Results One of the drawbacks of counting frequent sets is that the size of the answer can itself be too large to fit into memory. When the data has 1000 attributes, the number of subsets of small size (say 1,2,3,4) is large, and these sets are more likely to be frequent than ....

Z. Hu, W.-N. Chin, and M. Takeichi. Calculating a new data mining algorithm for market basket analysis. In Second International Workshop on Practical Aspects of Declarative Languages (PADL'00), Lecture Notes in 12 Computer Science, Boston, Massachusetts, January 2000. Springer Verlag.

....In this paper, we identify one important class of data mining problems called longest range problems, and propose a compositional framework for solving the problems e#ciently. This work is a continuation of our e#ort to investigate how program calculation approach could be used in data mining [HCT00], and it confirms us with its promising result. At present, our framework provide a wide class of predicates for specifying the range property, we want further to investigate how general predicates our approach can deal with. Another work we want to study is whether our framework can deal with ....

Z. Hu, W.N. Chin, and M. Takeichi. Calculating a new data mining algorithm for market basket analysis. In Second International Workshop on Practical Aspects of Declarative Languages, LNCS 1753, pages 169-- 184, Boston, Massachusetts, January 2000. Springer-Verlag.

....In this paper, we identify one important class of data mining problems called longest range problems, and propose a compositional framework for solving the problems e#ciently. This work is a continuation of our e#ort to investigate how program calculation approach could be used in data mining [HCT00], and it confirms us with its promising result. As a future work, we want to investigate how to extend our framework to mining multiple range attributes e#ciently. For detailed explanation and experimental results, please refer to [ZHM02a] ....

Z. Hu, W.N. Chin, and M. Takeichi. Calculating a new data mining algorithm for market basket analysis. In Proc. of PADL2000, LNCS 1753, pages 169--184, Boston, Massachusetts, January 2000. Springer-Verlag.

No context found.

Zhenjiang Hu, Wei-Ngan Chin, and Masato Takeichi. Calculating a New Data Mining Algorithm for Market Basket Analysis. Lecture Notes in Computer Science, 1753, 2000.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC