IARMMD: A Novel System for Incremental Association Rules Mining from Medical Documents

Abstract

This paper presents a novel system for Incremental Association Rules Mining from Medical Documents (IARMMD). The system concerns with maintenance of the discovered association rules and avoids redoing the mining process on whole documents during the updating process. The design of the system is based on concepts representation. It designed to develop our previous D-EART system. The IARMMD improves the updating process, and will lead to decrease the number of scanning and the execution time. The system consists of three phases that are Text Preprocessing, Incremental Association Rule Mining, and Visualization phase. Hash-based Incremental Association Rule Mining Algorithm (HIARM) is introduced in the mining phase. The algorithm employs the power of data structure called Hash Table. The performance of the algorithm is compared with both Apriori and FUP algorithms for the execution time and the evaluation of the extracted association rules. The results reveal that the number of extracted association rules in the IARMMD system is always less than that in Apriori-based and FUP-based systems. Furthermore, the execution time of HIARM algorithm is much better than Apriori and FUP algorithms in the updating process in all experimental cases.

Theoretical Analysis of Different Classifiers under Reduction Rough Data Set

“…Association rule mining proceeds on two main steps. The first step is to find all item sets with adequate supports and the second step is to generate association rules by combining these frequent or large item-sets (Mahgoub, 2006;Kannan & Bhaskaran, 2009). In the traditional association rules mining (Smith et al, 2004), minimum support threshold and minimum confidence threshold values are assumed to be available for mining frequent item sets, which is hard to be set without specific knowledge; users have difficulties in setting the support threshold to obtain their required results.…”

Survey on Apriori Algorithm for Market Analysis

“…Apriori is a seminal algorithm for finding frequent item-sets using candidate generation [9]. Mining for association among items in a large database of sales transaction is an important database mining function.…”

“…The first step is to find all item sets with adequate supports and the second step is to generate association rules by combining these frequent or large item-sets [9][10][l1].In the traditional association rules mining [12][13], minimum support threshold and minimum confidence threshold values are assumed to be available for mining frequent item sets, which is hard to be set without specific knowledge; users have difficulties in setting the support threshold to obtain their required results. Setting the support threshold too large, would produce only a small number of rules or even no rules to conclude.…”

Análise de 'backlog' de estórias de usuário por meio de agrupamento de textos e similaridade semântica

“…Regras de associação visam extrair correlações do tipo "se A então B", em que A e B são conjunções distintas de pares atributo-valor. De acordo com Mahgoub (2006), regras de associação no contexto de mineração de texto podem ser construídas com base em palavras-chaves e coleção de índices de documentos. Segundo Mahgoub (2006) ex.…”

“…De acordo com Mahgoub (2006), regras de associação no contexto de mineração de texto podem ser construídas com base em palavras-chaves e coleção de índices de documentos. Segundo Mahgoub (2006) ex. Decision Tree) e classificação por máquinas de vetores de suporte.…”