Enhancing the Apriori Algorithm for Frequent Set Counting
| Autor: | Paolo Palmerini, Salvatore Orlando, Raffaele Perego |
|---|---|
| Rok vydání: | 2001 |
| Předmět: |
Apriori algorithm
Association rule learning Computer science business.industry computer.software_genre Machine learning Set (abstract data type) Knowledge discovery Counting problem Knowledge extraction Data mining Artificial intelligence Pruning (decision trees) business Database Applications. Data mining computer FSA-Red Algorithm |
| Zdroj: | Data Warehousing and Knowledge Discovery, Third International Conference, pp. 71–82, Munich, Germany, 5-7 september 2001 info:cnr-pdr/source/autori:Orlando S.; Palmerini P.; Perego R./congresso_nome:Data Warehousing and Knowledge Discovery, Third International Conference/congresso_luogo:Munich, Germany/congresso_data:5-7 september 2001/anno:2001/pagina_da:71/pagina_a:82/intervallo_pagine:71–82 Data Warehousing and Knowledge Discovery ISBN: 9783540425533 DaWaK |
| Popis: | In this paper we propose DCP, a new algorithm for solving the Frequent Set Counting problem, which enhances Apriori. Our goal was to optimize the initial iterations of Apriori, i.e. the most time consuming ones when datasets characterized by short or medium length frequent patterns are considered. The main improvements regard the use of an innovative method for storing candidate set of items and counting their support, and the exploitation of effective pruning techniques which significantly reduce the size of the dataset as execution progresses. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|