A Matrix Approach for Association Mining

2003

Frequent itemset mining is a classic problem in data mining. It is a non-supervised process which concerns in finding frequent patterns (or itemsets) hidden in large volumes of data in order to produce compact summaries or models of the database. These models are typically used to generate association rules, but recently they have also been used in far reaching domains like e-commerce and bio-informatics. Because databases are increasing in terms of both dimension (number of attributes) and size (number of records), one of the main issues in a frequent itemset mining algorithm is the ability to analyze very large databases. Sequential algorithms do not have this ability, especially in terms of run-time performance, for such very large databases. Therefore, we must rely on high performance parallel and distributed computing. We present new parallel algorithms for frequent itemset mining. Their efficiency is proven through a series of experiments on different parallel environments, that range from shared-memory multiprocessors machines to a set of SMP clusters connected together through a high speed network.

Proceedings of the ninth ACM SIGKDD …, 2003

Existing association rule mining algorithms suffer from many problems when mining massive transactional datasets. One major problem is the high memory dependency: either the gigantic data structure built is assumed to fit in main memory, or the recursive mining process is too voracious in memory resources. Another major impediment is the repetitive and interactive nature of any knowledge discovery process. To tune parameters, many runs of the same algorithms are necessary leading to the building of these huge data structures time and again. This paper proposes a new disk-based association rule mining algorithm called Inverted Matrix, which achieves its efficiency by applying three new ideas. First, transactional data is converted into a new database layout called Inverted Matrix that prevents multiple scanning of the database during the mining phase, in which finding frequent patterns could be achieved in less than a full scan with random access. Second, for each frequent item, a relatively small independent tree is built summarizing co-occurrences. Finally, a simple and non-recursive mining process reduces the memory requirements as minimum candidacy generation and counting is needed. Experimental studies reveal that our Inverted Matrix approach outperform FP-Tree especially in mining very large transactional databases with a very large number of unique items. Our random access disk-based approach is particularly advantageous in a repetitive and interactive setting.

1997

Abstract Discovery of association rules is an important data mining task. Several parallel and sequential algorithms have been proposed in the literature to solve this problem. Almost all of these algorithms make repeated passes over the database to determine the set of frequent itemsets (a subset of database items), thus incurring high I/O overhead. In the parallel case, most algorithms perform a sum-reduction at the end of each pass to construct the global counts, also incurring high synchronization cost.

— The problems of mining association rules in a database are introduced. Most of association rule mining approaches aim to mine association rules considering exact matches between items in transactions. A new algorithm called " Improved Data Mining Based on Semantic Similarity to mine new Association Rules " which considers not only exact matches between items, but also the semantic similarity between them. Improved Data Mining (IDM) Based on Semantic Similarity to mine new Association Rules uses the concepts of an expert to represent the similarity degree between items, and proposes a new way of obtaining support and confidence for the association rules containing these items. Rapid advances in both data storage and data capture technologies have resulted in a marked increase in the amount of data being stored in the database of businesses, government agencies and scientific sectors. Because of these advances, millions of records are being generated and stored, each of them containing tens or hundreds of fields. Many of these datasets are expanding on a daily basis. Traditional analyses of these types of datasets, involving human experts who manually analyze the data, are clearly no longer adequate. Association rule mining [5] finds interesting associations and/or correlation relationships among large set of data items. It provides information in the form of "if-then" statements. Association rules shows attribute value conditions that occur frequently together in a given dataset. These rules are computed from the data and, unlike the if-then rules of logic, association rules are probabilistic in nature. A typical and widely used example of association rule mining is Market Basket Analysis. This paper Surveys some of the algorithms that have been introduced so far.

2012

One of the important problems in data mining is discovering association rules from databases of transactions where each transaction consists of a set of items. Many industries are interested in developing the association rules from their databases due to continuous retrieval and storage of huge amount of data. The discovery of interesting association relationship among business transaction records in many business decision making process such as catalog decision, cross-marketing, and loss-leader analysis. The most time consuming operation in this discovery process is the computation of the frequency of interesting subset of items (called candidates) in the database of transactions. Hence, it is has become vital to develop a method that may avoid or reduce candidate generation and test and utilize some novel data structures to reduce the cost in frequent pattern mining. In this paper, We have proposed An Integrated approach of Parallel Computing and ARM for mining Association Rules i...

A parallel algorithm for finding the frequent itemsets in a set of transactions is presented. The frequent individual items are identified by their index. We assume that processors number (m) is less than the frequent items number (n). At the first stage, every processor Pi, i isin; {1, ...,m - 1} sequentially computes the frequent itemsets from the interval Ii = [(i - 1) cdot; p + 1, i cdot; p], where p = lfloor;n/mrfloor;. The processor Pm computes frequent itemsets from the interval Im = [(m - 1) cdot; p + 1, n]. In the second stage, the parallel algorithm is applied. The processor Pi computes, step by step, the sets FIi,Ij of the frequent itemsets with individual items from the intervals Ii,j = Ii∪Ii+1∪...∪Ij, j = i+1,...,m. In order to compute the set FIi,Ij, the processor Pi uses FIi,Ij-1 obtained in the previous step and FIi+1,Ij received from the processor Pi+1. The main advantage of our parallel algorithm is that it uses a communication pattern known before algorithm start,...

Data Mining, 2001. ICDM 2001, …, 2001

Searching for frequent patterns in transactional databases is considered one of the most important data mining problems. Most current association mining algorithms, whether sequential or parallel, adopt an apriori-like algorithm that requires full multiple I/O scans of the data set and expensive computation to generate the potential frequent items. The recent explosive growth in data collection made the current association rule mining algorithms restricted and inadequate to analyze excessively large transaction sets due to the above mentioned limitations. In this paper we introduce a new parallel algorithm MLFPT (Multiple Local Frequent Pattern Tree) for parallel mining of frequent patterns, based on FP-growth mining, that uses only two full I/O scans of the database, eliminating the need for generating the candidate items, and distributing the work fairly among processors to achieve near optimum load balance. We have devised partitioning strategies at different stages of the mining process to achieve near optimal balancing between processors. This algorithm has been experimented on databases made of hundreds of thousands of dimensions and size larger than 10 Giga bytes using 64 processors SGI origin shared memory machine. We have successfully tested our algorithm on datasets larger than 50 million transactions. This paper presents the results of our algorithm on different data sizes experimented on different numbers of processors, and studies the effect of these variations on the overall performance.

Mining association rule is one of the key problems in data mining approach. Association rules discover the hidden relationships between various data items. In this paper, we propose a framework for the discovery of association rules using frequent pattern mining. We use preprocessing to transform the transaction dataset into a 2D matrix of 1's and 0's. Mining association rule must firstly discover frequent itemsets and then generate strong association rules from the frequent itemsets. The Apriori algorithm is the most well known association rule mining algorithm and is less efficient because they need to scan the database many times and store transaction ID in memory, so time and space overhead is very high. Especially they are less efficient when they process large scale database. Here we propose improved Apriori algorithm by including prune step and hash map data structure. The improved algorithm is more suitable for large scale database. Experimental results shows that computation times are reduced by using the prune step and hash map data structure.

International Journal of Future Computer and Communication, 2013

Advances in Pattern …, 2010

Abstract. Mining frequent itemsets in large databases is a widely used technique in Data Mining. Several sequential and parallel algorithms have been developed, although, when dealing with high data volumes, the execution of those algorithms takes more time and resources ...