Zailani Abdullah

Mining significant least association rules using fast SLP-growth algorithm

Development of least association rules mining algorithms are very challenging in data mining. The complexity and excessive in computational cost are always become the main obstacles as compared to mining the frequent rules. Indeed, most of the previous studies still adopting the Apriori-like algorithms which are very time consuming. To address this issue, this paper proposes a scalable trie-based algorithm named SLP-Growth. This algorithm generates the significant patterns using interval support and determines its correlation. Experiments with the real datasets show that the SLP-algorithm can discover highly positive correlated and significant of least association. Indeed, it also outperforms the fast FP-Growth algorithm up to two times, thus verifying its efficiency.

Mining Interesting Association Rules of Student Suffering Mathematics Anxiety

Up to this moment, association rules mining are one of the most important issues in data mining application. One of the commonly and popular techniques used in data mining application is association rules mining. The purpose of this study is to apply an enhanced association rules mining method, so called SLP-Growth (Significant Least Pattern Growth) proposed by [9] for capturing interesting rules in student suffering mathematics anxiety dataset. The dataset was taken from a survey on exploring mathematics anxiety among engineering students in Universiti Malaysia Pahang (UMP). The results of this research will provide useful information for educators to make a decision on their students more accurately, and to adapt their teaching strategies accordingly. It also can be helpful to assist students in handling their fear of mathematics and useful in increasing the quality of learning.

Scalable model for mining critical least association rules

A research in mining least association rules is still outstanding and thus requiring more attentions. Until now; only few algorithms and techniques are developed to mine the significant least association rules. In addition, mining such rules always suffered from the high computational costs, complicated and required dedicated measurement. Therefore, this paper proposed a scalable model called Critical Least Association Rule (CLAR) to discover the significant and critical least association rules. Experiments with a real and UCI datasets show that the CLAR can generate the critical least association rules, up to 1.5 times faster and less 100% complexity than benchmarked FP-Growth.

An Alternative Measure for Mining Weighted Least Association Rule and Its Framework

Mining weighted based association rules has received a great attention and consider as one of the important area in data mining. Most of the items in transactional databases are not always carried with the same binary value. Some of them might associate with different level of important such as the profit margins, weights, etc. However, the study in this area is quite complex and thus required an appropriate scheme for rules detection. Therefore, this paper proposes a new measure called Weighted Support Association Rules (WSAR*) measure to discover the significant association rules and Weighted Least Association Rules (WELAR) framework. Experiment results shows that the significant association rules are successfully mined and the unimportant rules are easily differentiated. Our algorithm in WELAR framework also outperforms the benchmarked FP-Growth algorithm.

Visualizing the construction of incremental disorder Trie Itemset data structure (DOSTrieIT) for frequent pattern tree (FP-tree)

In data mining, visual representation can help in enhancing the ability of analyzing and understanding the techniques, patterns and their integration. Recently, there are varieties of visualizers have been proposed in marketplace and knowledge discovery communities. However, the detail visualization processes for constructing any incremental tree data structure from its original dataset are rarely presented. Essentially, graphic illustrations of the complex processes are easier to be understood as compared to the complex computer pseudocode. Therefore, this paper explains the visualization process of constructing our incremental Disorder Support Trie Itemset (DOSTrieIT) data structure from the flat-file dataset. DOSTrieIT can be used later as a compressed source of information for building Frequent Pattern Tree (FP-Tree). To ensure understandability, an appropriate dataset and its processes are graphically presented and details explained.

CNAR-M: A Model for Mining Critical Negative Association Rules

Association rules mining has been extensively studied in various multidiscipline applications. One of the important categories in association rule is known as Negative Association Rule (NAR). Significant NAR is very useful in certain domain applications; however it is hardly to be captured and discriminated. Therefore, in this paper we proposed a model called Critical Negative Association Rule Model (CNAR-M) to extract the Critical Negative Association Rule (CNAR) with higher Critical Relative Support (CRS) values. The result shows that the CNAR-M can mine CNAR from the benchmarked and real datasets. Moreover, it also can discriminate the CNAR with others association rules.

Mining highly correlated least association rules using scalable trie-based algorithm

Development of least association rules (ARs) mining algorithms is one of the more challenging areas in data mining. Exclusive measurements, complexity and excessive computational cost are the main obstacles as compared to frequent pattern mining. Indeed, most previous studies still use the Apriori-like algorithms. To address this issue, this article proposes a new correlation measurement called definite factor (DF) and a scalable trie-based algorithm named significant least pattern growth (SLP-Growth). This algorithm generates the least patterns based on interval support and finally determines it significances using DF. Experiments with the real datasets show that the SLP-Growth can discover highly positive correlated and significant of least ARs. Indeed, it also outperforms the fast frequent pattern-Growth algorithm up to two times, thus verifying its efficiency.

DETECTING CRITICAL LEAST ASSOCIATION RULES IN MEDICAL DATABASES

Least association rules are corresponded to the rarity or irregularity relationship among itemset in database. Mining these rules is very difficult and rarely focused since it always involves with infrequent and exceptional cases. In certain medical data, detecting these rules is very critical and most valuable. However, mathematical formulation and evaluation of the new proposed measurement are not really impressive. Therefore, in this paper we applied our novel measurement called Critical Relative Support (CRS) to mine the critical least association rules from medical dataset. We also employed our scalable algorithm called Significant Least Pattern Growth algorithm (SLP-Growth) to mine the respective association rules. Experiment with two benchmarked medical datasets, Breast Cancer and Cardiac Single Proton Emission Computed Tomography (SPECT) Images proves that CRS can be used to detect to the pertinent rules and thus verify its scalability.

Mining Interesting Association Rules of Students Suffering Study Anxieties Using SLP-Growth Algorithm

One of the most popular techniques used in data mining applications is association rules mining. The purpose of this study is to apply an enhanced association rules mining method, called SLP-Growth Significant Least Pattern Growth for capturing interesting rules from students suffering mathematics and examination anxieties datasets. The datasets were taken from a survey exploring study anxieties among engineering students in Universiti Malaysia Pahang UMP. The results of this research provide useful information for educators to make decisions on their students more accurately and adapt their teaching strategies accordingly. It also can assist students in handling their fear of mathematics and examination and increase the quality of learning.

IPMA : indirect patterns mining algorithm

Indirect pattern is considered as valuable and hidden information in transactional database. It represents the property of high dependencies between two items that are rarely occurred together but indirectly appeared via another items. Indirect pattern mining is very important because it can reveal a new knowledge in certain domain applications. Therefore, we propose an Indirect Pattern Mining Algorithm (IPMA) in an attempt to mine the indirect patterns from data repository. IPMA embeds with a measure called Critical Relative Support (CRS) measure rather than the common interesting measures. The result shows that IPMA is successful in generating the indirect patterns with the various threshold values.