2011Unpublished venueRequires access

An improved multi-support Apriori algorithm under the fuzzy item association condition

Hangbin Li, Shuhua Chen, LI Jianchen, Shuo Wang, Yihang Fu

Open publisher page 1 citations

Abstract

Apriori Algorithm has been successfully used in the mining of association rules. The algorithm with multiple-support is proposed for solving "rare item" problem, which means the common items and rare items cannot be satisfied within single support. However, the detailed identification of items will cause the poor mining rules, while the multi-level approach distinguishes the fuzzy association between items in real condition. This paper proposed a novel model based on tags to solve the problem. Each item has various tags, which implies the common attributes. This model is adaptive to the reality since certain item set can be accurately derived from whole sets using combination of tags by users as they like. A corresponding improved multi-support Apriori algorithm is also presented for a recommendation system. The experimental results demonstrated that the new model and the approach are more effective and adaptive.

About this research paper

What this paper is about

Apriori Algorithm has been successfully used in the mining of association rules. The algorithm with multiple-support is proposed for solving "rare item" problem, which means the common items and rare items cannot be satisfied within single support. However, the detailed identification of items will cause the poor mining rules, while the multi-level approach distinguishes the fuzzy association between items in real condition. This paper proposed a novel model based on tags to solve the problem. Each item has various tags, which implies the common attributes. This model is adaptive to the reality since certain item set can be accurately derived from whole sets using combination of tags by users as they like. A corresponding improved multi-support Apriori algorithm is also presented for a recommendation system. The experimental results demonstrated that the new model and the approach are more effective and adaptive.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Apriori Algorithm has been successfully used in the mining of association rules. The algorithm with multiple-support is proposed for solving "rare item" problem, which means the common items and rare items cannot be satisfied within single support. However, the detailed identification of items will cause the poor mining rules, while the multi-level approach distinguishes the fuzzy association between items in real condition. This paper proposed a novel model based on tags to solve the problem. Each item has various tags, which implies the common attributes. This model is adaptive to the reality since certain item set can be accurately derived from whole sets using combination of tags by users as they like. A corresponding improved multi-support Apriori algorithm is also presented for a recommendation system. The experimental results demonstrated that the new model and the approach are more effective and adaptive.

Key concepts: Apriori algorithm, Association rule learning, Computer science, A priori and a posteriori, Data mining, Identification (biology), Set (abstract data type), Fuzzy set

Related papers

Back to paper searchBrowse research topicsOriginal source
An improved multi-support Apriori algorithm under the fuzzy item association condition — Research Paper | ScholarLens