Wyniki wyszukiwania - BazTech

Ograniczanie wyników

2 Systems Science

Powiadomienia systemowe

Sesja wygasła!
Sesja wygasła!

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

The other thresholds in the mining association algorithm.

Kawahara M., Kawano H

Systems Science

2000

Vol. 26, nr 2

95-109

The mining association algorithm is one of the most important data mining algorithms to derive association rules at high speed from huge databases. However, the rules derived by the algorithm contain many noises; it is then necessary for some systems to remove the noises using noise filters such as stopword lists. We improve the algorithm and develop bibliographic navigation systems using the algorithms, and we also use a dictionary to remove noises in association rules. In order to derive effective rules, it is very important to determine system parameters, such as Minsup and Minconf threshold values. We adapt the receiver operating characteristic analysis on the algorithm in our bibliographic navigation system and evaluate the performance of the derived rules. We propose extended mining association algorithms. Moreover, we evaluate the performance of our algorithms proposed using experimental database and show how our algorithms can derive effective association rules. We also show that our algorithm can remove stopwords from raw data automatically.

Implementation of bibliographic navigation system with text data mining.

Kawahara M., Kawano H., Hasegawa T.

Systems Science

1998

Vol. 24, nr 3

101-118

It is generally difficult for naive users to retrieve appropriate bibliographies from bibliographic databases, even if they have background and domain knowledge. One of the reason of the difficulty is that value of the attribute is not limited to a certain values. In this paper, we describe an implementation of our bibliographic navigation system with the mining associated algorithm which was extended by us from one of the data mining algorithm. In order to provide more helpful knowledge, we categorized the attributes and extend the mining association algorithms which discover relevance to a keywords space. By our algorithm, interesting rules are derived from relationships between several categorized attributes in a database. We have verified the performance of the proposed algorithms on our practical bibliographic navigation system.