R. C. Agarwal, C. C. Aggarwal, and V.V.V. Prasad. Depth first generation of long patterns. In Proc. of the 6th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, pages 108--118, 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....the pattern discovery algorithm we developed earlier. We then use these patterns to classify the proteins. Experimental results show the good performance of the proposed approach. 1 Introduction Discovering frequently occurring patterns has been explored in many di#erent domains, e.g. sequences [1], trees [6] semistructured data [7] three dimensional data [9] Classification is also one of the major tasks of data mining [3] Protein classification is a very important research topic [3, 4, 6] Traditionally, proteins are classified according to their functions. However, recently, many ....

R. C. Agarwal, C. C. Aggarwal, and V. V. V. Prasad, "Depth first generation of long patterns, " Proceedings of the 6th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 108--118, Boston, Massachusetts, 2000.

....a bottomup manner like Apriori, but also starts a top down search at the same time. This can help in reducing the number of database scans. MaxMiner [4] is another algorithm for finding the maximal elements. It uses e# cient pruning based on lookaheads to quickly narrow the search. DepthProject [2] finds long itemsets using a depth first search of a lexicographic tree of itemsets, and uses a counting method based on transaction projections along its branches. Mafia [6] also uses several pruning strategies, uses vertical bit vector data format, and compression and projection of bitmaps to ....

....the vertical format maintains for each item its tidset, a set of all tids where it occurs. Most of the past research has utilized the traditional horizontal database format for mining; some of these methods include Apriori [1] that mines frequent itemsets, and MaxMiner [4] and DepthProject [2] which mine maximal itemsets. Notable exception to this trend are the approaches that A A A C T W C D W C T W C D W C D T W C D T 1 3 4 5 6 5 3 4 2 1 4 2 5 6 6 5 3 1 1 2 3 4 5 A C D T W 1 1 1 A C D T W 1 1 1 1 1 4 5 6 3 2 1 1 1 1 1 1 1 1 1 1 1 1 ....

Ramesh Agrawal, Charu Aggarwal, and V.V.V. Prasad. Depth First Generation of Long Patterns. In 7th Int'l Conference on Knowledge Discovery and Data Mining, August 2000.

....to compactly store in memory the itemsets of the original database. The basic ideas in this algorithm were recently used to develop a similar algorithm for finding sequential patterns [18] The problem of finding frequent patterns has been extended to that of finding frequent maximal patterns [4, 10, 2, 27] and finding frequent closed patterns [15, 23, 17] Both of these problem formulations can be used to reduce the number of patterns that gets discovered and help in finding long patterns present in the data. However, both of these problem formulations can still generate a very large number of ....

Ramesh C. Agarwal, Charu C. Aggarwal, and V. V. V. Prasad. Depth first generation of long patterns. In Knowledge Discovery and Data Mining, pages 108-118, 2000.

....that have minimum support, because this operation is by far the most expensive phase of the mining process. We assume that the reader is familiar with the basic Apriori algorithm introduced in [3] A variety of modifications have been proposed to reduce the computational burden see, for example, [2, 5, 7, 14] and references therein but with few exceptions all current algorithms require at least one expensive pass over the data. In the context of standard association rule mining, use of samples can make mining studies feasible that were for merly impractical due to the enormous time requirements. ....

....an overall cost that is much lower than that of classical algorithms. We emphasize that the sample created by fast can be subsequently processed by any existing (non sampling based) association rule algorithm, so that fast complements, rather than replaces, current algorithms such as DepthMiner [2], Max Miner [5] DIC [7] or FP tree [14] The fast technique is especially compatible with algorithms such as DepthMiner that are designed for memory resident data sets. Moreover, our new sampling technique can potentially be applied to render scalable other mining and statistical ....

R. C. Agarwal, C. C. Aggarwal, and V. V. V. Prasad. Depth first generation of long patterns. In Proc. Sixth ACM SIGKDD, 2000.

....condensed FP base w.r.t. function but also a minimal one. Limited by space, we omit the formal result here. The remaining problem is how to find the max patterns efficiently in the condensed FP base . There are many methods for mining max patterns, such as MaxMiner [3] Depth first Search [1], MAFIA [5] and GenMax[7] A na ve method to compute is to call a max pattern mining algorithm multiple times, once for each lower bound of the ranges as a support threshold. How do we mine the patterns of M bases more efficiently than the na ve method Roughly speaking, we will propose an ....

R.C. Agarwal, C.C. Aggarwal, V. V. V. Prasad. Depth first generation of long patterns. In KDD'00.

....to compactly store in memory the itemsets of the original database. The basic ideas in this algorithm were recently used to develop a similar algorithm for finding sequential patterns [18] The problem of finding frequent patterns has been extended to that of finding frequent maximal patterns [4, 10, 2, 27] and finding frequent closed patterns [15, 23, 17] Both of these problem formulations can be used to reduce the number of patterns that gets discovered and help in finding long patterns present in the data. However, both of these problem formulations can still generate a very large number of ....

Ramesh C. Agarwal, Charu C. Aggarwal, and V. V. V. Prasad. Depth first generation of long patterns. In Knowledge Discovery and Data Mining, pages 108--118, 2000.

No context found.

R. C. Agarwal, C. C. Aggarwal, and V.V.V. Prasad. Depth first generation of long patterns. In Proc. of the 6th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, pages 108--118, 2000.

No context found.

R. C. Agarwal, C. C. Aggarwal, and V.V.V. Prasad. Depth first generation of long patterns. In Proc. of the 6th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, pages 108--118, 2000.

No context found.

R.C. Agarwal, C.C. Aggarwal, and V.V.V. Prasad. Depth first generation of long patterns. In Ramakrishnan et al. [32], pages 108--118.

No context found.

R. C. Agarwal, C. C. Aggarwal, and V. V. V. Prasad. Depth first generation of long patterns. In Proceedings of the ACM SIGKDD Conference, 2000.

No context found.

R. C. Agarwal, C. C. Aggarwal, and V. Parsad. Depth first generation of long patterns. In SIGKDD, 2000.

No context found.

Agarwal, R. C., Aggarwal, C. C., and Prasad, V. V. Depth first generation of long patterns. In SIGKDD

No context found.

R.C. Agarwal, C.C. Aggarwal, and V.V.V. Prasad. Depth first generation of long patterns. In Ramakrishnan et al. [69], pages 108--118.

No context found.

R.C. Agarwal, C.C. Aggarwal, and V.V.V. Prasad. Depth first generation of long patterns. In Proceedings of the Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp 108--118. Boston, MA, USA, August 2000.

No context found.

Agarwal,R.C.,Aggarwal, C. C., and Prasad, V. V. Depth first generation of long patterns. In SIGKDD

No context found.

R. C. Agarwal, C. C. Aggarwal, and V. V. V. Prasad. Depth first generation of long patterns. In Knowledge Discovery and Data Mining, pages 108--118, 2000.

No context found.

R. C. Agarwal, C. C. Aggarwal, and V. V. V. Prasad. Depth first generation of long patterns. In Knowledge Discovery and Data Mining, pages 108--118, 2000.

No context found.

R. C. Agarwal, C. C. Aggarwal, and V.V.V. Prasad. Depth first generation of long patterns. In Proc. of the 6th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, pages 108--118, 2000.

No context found.

R. C. Agarwal, C. C. Aggarwal, and V.V.V. Prasad. Depth first generation of long patterns. In Proc. of the 6th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, pages 108--118, 2000.

No context found.

R. C. Agarwal, C. C. Aggarwal, V. Prasad. Depth First Generation of Long Patterns. In Proc. of KDD 2000, Boston, USA.

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