V. Pudi and J. Haritsa. Quantifying the utility of the past in mining large databases. Information Systems, July 2000. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
A Framework for Incremental Generation of Frequent Closed.. - Valtchev, Missaoui, Godin (2002) (Correct)
.... Although the necessity of processing volatile data in an incremental manner have been repeatedly emphasized in the general data mining literature (see for example [9] a few incremental algorithms for association rule generation (and hence frequent itemset detection) have been reported so far [5, 6, 15]. The conclusions drawn from this studies poited at an aditional storage demand due to the impossibility to prune some of the infrequent itemsets in run time. Our own approach for incremental FI generation is based on the intuitive idea that FCIs may be the right answer to the problem of the ....
....association rules when insertion, deletion, and modification of transactions occur. Both FUP and FUP2 are based on the Apriori framework (e.g. there is a candidate generation step) that exploits the previous mining output to avoid the generation of useless candidates. A recent work reported in [15] extends the limits of incremental approaches by allowing changes to the basic parameters of the mining process (e.g. minsupp threshold, number of transactions added at a time, etc. Alternative approaches to mining CIs from a database have been presented in [21, 13] both following the ....
[Article contains additional citation context not shown here]
V. Pudi and J. R. Haritsa. Quantifying the Utility of the Past in Mining Large Databases. Information Systems, 25(5):323-343, 2000.
On the Efficiency of Association-rule Mining Algorithms - Pudi, Haritsa (2002) Self-citation (Pudi Haritsa) (Correct)
No context found.
V. Pudi and J. Haritsa. Quantifying the utility of the past in mining large databases. Information Systems, July 2000.
ARMOR: Association Rule Mining based on ORacle - Vikram Pudi Intl Self-citation (Pudi Haritsa) (Correct)
No context found.
V. Pudi and J. Haritsa. Quantifying the utility of the past in mining large databases. Information Systems, July 2000.
ARMOR: Association Rule Mining based on ORacle - Pudi, Haritsa Self-citation (Pudi Haritsa) (Correct)
No context found.
V. Pudi and J. Haritsa. Quantifying the utility of the past in mining large databases. Information Systems, July 2000.
On the Efficiency of Association-rule Mining Algorithms - Pudi, Haritsa (2002) Self-citation (Pudi Haritsa) (Correct)
No context found.
V. Pudi and J. Haritsa. Quantifying the utility of the past in mining large databases. Information Systems, July 2000.
On the Optimality of Association-rule Mining Algorithms - Pudi, Haritsa (2001) Self-citation (Pudi Haritsa) (Correct)
....Mining based on ORacle) whose structure is derived by making minimal changes to the Oracle, and is guaranteed to complete in two passes over the database. ARMOR incorporates techniques from a variety of previous algorithms such as PARTITION [20] CARMA [13] AS CPA [15] VIPER [21] and DELTA [19]. Our empirical study shows 1 that ARMOR performs within a factor of two of the Oracle, over a variety of databases and practical ranges of support specifications. Finally, an important feature of our experiments is that they include workloads where the database is large enough that the working ....
....that computes the set of frequent itemsets and accesses the data using only queries of the following form: Is itemset X frequent must use at least jN j such queries. ffl The negative border information has been found to be especially useful in the design of incremental mining algorithms [19, 24, 10]. These algorithms are designed to efficiently derive the current mining output by utilizing previous mining results when a database has been updated with an increment. For ease of exposition, we will use the notation shown in Table 1 in the remainder of this paper. Mining Algorithms ....
[Article contains additional citation context not shown here]
V. Pudi and J. Haritsa. Quantifying the utility of the past in mining large databases. Information Systems, July 2000.
Quantifying the Utility of the Past in Mining Large Databases - Pudi, Haritsa (2000) (1 citation) Self-citation (Pudi Haritsa) (Correct)
....prunes away most of the unnecessary candidates in its second pass over the increment. Due to space constraints, the experimental results for hierarchical databases where the increment s distribution is Skewed, as also the multi support environments, are not presented here. They are available in [15] and are similar in nature to those presented earlier in this paper for flat databases. 7. CONCLUSIONS We considered the problem of incrementally mining association rules on market basket databases that have been subjected to a significant number of updates since their previous mining exercise. ....
V. Pudi and J. Haritsa. Quantifying the utility of the past in mining large databases. Technical Report TR-2000-01, DSL, Indian Institute of Science, http://dsl.serc.iisc.ernet.in/pub/TR/TR-2000-01.ps (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