Heikki Mannila and HannuToivonen. On an algorithm for finding all interesting sentences extended abstract. In Cybernetics and Systems Research, 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....database can be fed into the memory, rather than scanning the large database. This method is extended in the distributed database, whichwas done in [9] 8] In this case, only O(n) messages, rather than O(n 2 ) messages, is needed for exchanging the counts. Other related work can be found in [38], which is the theoretical proof of this rule. The visualization of the rule is done in [33] and the multi dimensional association rules by using the data cubes is done by[24] In summary, the database oriented approach can provide a straight and easy way to mine the information from the ....

Heikki Mannila and HannuToivonen. On an algorithm for finding all interesting sentences extended abstract. In Cybernetics and Systems Research, 1996.

....: X k ) interest(X 1 ; X k ) ffl Chapter 4 The ENCLUS Algorithms 39 Hence, the subspace X 1 ; X n is also correlated. 2 4.3 Complexity Analysis In this section, we examine the worst case complexity of our algorithm. Note that the target problem of Apriori is NP hard [27], so an algorithm based on the same framework is unlikely to run in polynomial time theoretically. However, we hope the algorithm would run faster than its theoretical bound in practice. The performance evaluation of the algorithm by experiment is presented in Chapter 5. First, we present the ....

Heikki Mannila and Hannu Toivonen. On an algorithm for finding all interesting sentences extended abstract. In Proceedings of the 6th Internation Conference on Database Theory, pages 215--229, 1996.

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