Fuzzy Multiple Support Associative Classification Approach for Prediction

Bilal Sowan, Keshav Dahal, Alamgir Hussain

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

The fact of building an accurate classification and prediction system remains one of the most significant challenges in knowledge discovery and data mining. In this paper; a Knowledge Discovery (KID) framework is proposed; based on the integrated fuzzy approach, more specifically Fuzzy C-Means (FCM) and the new Multiple Support Classification Association Rules (MSCAR) algorithm. MSCAR is considered as an efficient algorithm for extracting both rare and frequent rules using vertical scanning format for the database. Consequently; the adaptation of such a process sufficiently minimized the prediction error. The experimental results regarding two data sets; Abalone and road traffic, show the effectiveness of the proposed approach in building a robust prediction system. The results also demonstrate that the proposed KID framework outperforms the existing prediction systems.
Original languageEnglish
Title of host publicationArtificial Intelligence and Soft Computing, Pt I
PublisherSpringer-Verlag
Pages216-223
Volume6113
ISBN (Print)978-3-642-13207-0
DOIs
Publication statusPublished - 2010
Externally publishedYes

Publication series

NameLecture Notes in Artificial Intelligence

Keywords

  • Knowledge Discovery
  • MSapriori
  • Apriori
  • Fuzzy C-Means
  • Classification
  • Prediction

Cite this

Sowan, B., Dahal, K., & Hussain, A. (2010). Fuzzy Multiple Support Associative Classification Approach for Prediction. In Artificial Intelligence and Soft Computing, Pt I (Vol. 6113, pp. 216-223). (Lecture Notes in Artificial Intelligence). Springer-Verlag. https://doi.org/10.1007/978-3-642-13208-7_28