Discovery of Maximal Frequent Item Sets using Subset Creation

Association rule has been an area of active research in the field of knowledge discovery. Data mining researchers had improved upon the quality of association rule mining for business development by incorporating influential factors like value (utility), quantity of items sold (weight) and more for the mining of association patterns. In this paper, we propose an efficient approach to find maximal frequent item set first. Most of the algorithms in literature used to find minimal frequent item first, then with the help of minimal frequent item sets derive the maximal frequent item sets. These methods consume more time to find maximal frequent item sets. To overcome this problem, we propose a navel approach to find maximal frequent item set directly using the concepts of subsets. The proposed method is found to be efficient in finding maximal frequent item sets.

In the discipline of data mining, association rule mining is an important study topic that focuses on discovering the relationships between database attributes. The maximum frequent itemset comprises the information of all frequent itemsets, which is one of the important difficulties in mining association rules, and certain data mining applications just need to mine the maximum frequent itemsets. As a result, analyzing the maximum frequent itemset mining technique is practical. Considering this, the research introduces FP-MFIA, a new maximum frequent itemset mining approach based on the FP-tree, which is inspired by the data structure of the frequent pattern tree and the idea that the maximum frequent itemset implies all frequent itemsets. First, the FP-MFIA constructs a one-way FP-tree structure, which only has pointers from the root to the leaves, so that only two scans of the FP-tree are required by the FP-MFIA. On the other hand, it redefines a data storage structure MFI-list for maximum frequent itemsets. It can quickly release unnecessary nodes in the FP-tree after scanning it. In this way, not only the information required by the maximum frequent itemsets can be quickly mined but also the space required for storing the maximum frequent itemsets can be reduced, which greatly improves the mining efficiency. Finally, experiments were conducted to compare the mining efficiency of the novel FP-MFIA algorithm to the IDMFIA and DMFIA algorithms. We can see from the findings that the FP-MFIA algorithm is more efficient than the other two techniques.

The association rule mining can be divided into two steps. The first step is to find out all frequent itemsets, whose occurrences are greater than or equal to the user-specified threshold. The second step is to generate reliable association rules based on all frequent itemsets found in the first step. Identifying all frequent itemsets in a large database dominates the overall performance in the association rule mining. In this paper, we propose an efficient hash-based method, HMFS, for discovering the maximal frequent itemsets. The HMFS method combines the advantages of both the DHP (Direct Hashing and Pruning) and the Pincer-Search algorithms. The combination leads to two advantages. First, the HMFS method, in general, can reduce the number of database scans. Second, the HMFS can filter the infrequent candidate itemsets and can use the filtered itemsets to find the maximal frequent itemsets. These two advantages can reduce the overall computing time of finding the maximal frequent itemsets. In addition, the HMFS method also provides an efficient mechanism to construct the maximal frequent candidate itemsets to reduce the search space. We have implemented the HMFS method along with the DHP and the Pincer-Search algorithms on a Pentium III 800 MHz PC. The experimental results show that the HMFS method has better performance than the DHP and the Pincer-Search algorithms for most of test cases. In particular, our method has significant improvement over the DHP and the Pincer-Search algorithms when the size of a database is large and the length of the longest itemset is relatively long.

Discovering frequent itemsets is a key problem in important data mining applications, such as the discovery of association rules, strong rules, episodes, and minimal keys. Typical algorithms for solving this problem operate in a bottom-up breadth-first search direction. The computation starts from frequent 1-itemsets (minimal length frequent itemsets) and continues until all maximal (length) frequent itemsets are found. During the execution, every frequent itemset is explicitly considered. Such algorithms perform reasonably well when all maximal frequent itemsets are short. However, performance drastically decreases when some of the maximal frequent itemsets are relatively long. We present a new algorithm which combines both the bottom-up and top-down directions. The main search direction is still bottom-up but a restricted search is conducted in the top-down direction. This search is used only for maintaining and updating a new data structure we designed, the maximum frequent candidate set. It is used to prune candidates in the bottom-up search. As a very important characteristic of the algorithm, it is not necessary to explicitly examine every frequent itemset. Therefore it performs well even when some maximal frequent itemsets are long. As its output, the algorithm produces the maximum frequent set, i.e., the set containing all maximal frequent itemsets, which therefore specifies immediately all frequent itemsets. We evaluate the performance of the algorithm using a well-known benchmark database. The improvements can be up to several orders of magnitude, compared to the best current algorithms.

When it comes to association rule mining, all frequent itemsets are first found, and then the confidence level of association rules is calculated through the support degree of frequent itemsets. As all non-empty subsets in frequent itemsets are still frequent itemsets, all frequent itemsets can be acquired only by finding all maximal frequent itemsets (MFIs), whose supersets are not frequent itemsets. In this study, an algorithm, named right-hand side expanding (RHSE), which can accurately find all MFIs, was proposed. First, an Expanding Operation was designed, which, starting from any given frequent itemset, could add items using certain rules and form some supersets of given frequent itemsets. In addition, these supersets were all MFIs. Next, this operator was used to add items by taking all frequent 1-itemsets as the starting point alternately, and all MFIs were found in the end. Due to the special design of the Expanding Operation, each MFI could be found. Moreover, the path found was unique, which avoided the algorithm redundancy in temporal and spatial complexity. This algorithm, which has a high operating rate, is applicable to the big data of high-dimensional mass transactions as it is capable of avoiding the computing redundancy and finding all MFIs. In the end, a detailed experimental report on 10 open standard transaction sets was given in this study, including the big data calculation results of million-class transactions.

One of the recent research topics in databases is Data Mining, to find, extract and mine the useful information from databases. In case of updating transactions in the database the already discovered knowledge may become invalid. So we need efficient knowledge management techniques for finding the updated knowledge from the database. There have been lot of research in data mining, but Knowledge Management in databases is not studied much. One of the data mining techniques is to find association rules from databases. But most of association rule algorithms find association rules from transactional databases. Our research is a further step of the Tree Based Association Rule Mining (TBAR) algorithm, used in relational databases for finding the association rules .In our approach of updating the already discovered knowledge; the proposed algorithm Association Rule Update (ARU), updates the already discovered association rules found through the TBAR algorithm. Our algorithm will be able to find incremental association rules from relational databases and efficiently manage the previously found knowledge.

Frequent itemsets (FIs) mining from uncertain database is a very popular research area nowadays. Many algorithms have been proposed to mine FI from uncertain database. But in typical FI mining process, all the FIs have to be mined individually, which needs a huge memory. Four trees are proposed in this paper which are: (i) maximal frequent itemset from uncertain database (MFU) tree which contains only the maximal frequent itemsets generated from uncertain database, (ii) closed frequent itemset from uncertain database (CFU) tree which contains only closed frequent itemsets generated from uncertain database, (iii) maximal frequent itemset from uncertain data stream (MFUS) tree which contains maximal frequent itemsets generated from uncertain data stream and (iv) closed frequent itemset from uncertain data stream (CFUS) tree which contains closed frequent itemsets generated from uncertain data stream. Experimental results are also presented which show that maximal and closed frequent itemsets mining requires less time and memory than typical frequent itemsets mining.

Association rule mining is an important topic in data mining research. Many algorithms have been developed for such task and they typically assume that the underlying associations hidden in the data are stable over time. However, in real world domains, it is possible that the data characteristics and hence the associations change significantly over time. Existing data mining algorithms have not taken the changes in associations into consideration and this can result in severe degradation of performance, especially when the discovered association rules are used for classification (prediction). Although the mining of changes in associations is an important problem because it is common that we need to predict the future based on the historical data in the past, existing data mining algorithms are not developed for this task. In this paper, we introduce a new fuzzy data mining technique to discover changes in association rules over time. Our approach mines fuzzy rules to represent the changes in association rules. Based on the discovered fuzzy rules, our approach is able to predict how the association rules will change in the future. The experimental results on a real-life database have shown that our approach is very effective in mining and predicting changes in association rules over time.

Data mining involves the use of advanced data analysis tools to find out new, suitable patterns and project the relationship among the patterns which were not known prior. In data mining, association rule learning is a popular and well researched method for discovering interesting relations between variables in large databases. One of the emerging research areas under Data mining is Social Networks. The objective of this paper focuses on the formulation of association rules using which decisions can be made for future Endeavour. This research applies Apriori Algorithm which is one of the classical algorithms for deriving association rules. The Algorithm is applied to Face book 100 university dataset which has originated from Adam D’Angelo of Face book. It contains self-defined characteristics of a person including variables like residence, year, and major, second major, gender, school. This paper to begin with the research uses only ten Universities and highlights the formation of association rules between the attributes or variables and explores the association rule between a course and gender, and discovers the influence of gender in studying a course. The previous research with this dataset has applied only regression models and this is the first time to apply association rules.

Maximal frequent item sets are one of several condensed representations of frequent item sets, which store most of the information contained in frequent item sets using less space, thus being more suitable for stream mining. This paper focuses on mining maximal frequent item sets approximately over a stream landmark model. We separate the continuously arriving transactions into sections, and the mining results are indexed by an extended direct update tree, thus, a simple but effective algorithm named SMIS is proposed. In our algorithm, we employ the Chern off Bound to perform the maximal frequent item set mining in a false negative manner, which can reduce the memory cost, as well guarantee our algorithm conducting with an incremental fashion. Our experimental results on two synthetic datasets and two real world datasets show that SMIS achieves much reduced memory cost in comparison with the state-of-the-art algorithm with a 100 percent precision.

Frequent itemset mining and association rule mining are fundamental problems in data mining. Despite of the intensive and continuous researches on frequent itemset mining, one essential and not completely solved drawback still remains. The drawback is pattern explosion: a huge number of frequent itemsets, or patterns, will be often derived. The association rule mining takes over the same drawback since each association rule consists of two frequent itemsets. One promising solution for alleviating this essential drawback practically as a post-processing is to rank the patterns and rules from various aspects and to identify which one to be examined first. In this paper, we propose to utilize the technology of representation learning to rank frequent itemsets and association rules. More specifically, we develop several evaluation measures for frequent itemsets and association rules by using the vector representations of items, transactions and frequent patterns which are obtained by existing representation learning techniques. By employing vector representations of three kinds of objects, various rankings reflecting different aspects can expect to be obtained.

Data mining aims to develop a new generation of tools to intelligently assist humans in analyzing mountains of data. Over the past few years, great progress has been made in both research and applications of data mining. Data mining systems have helped many businesses by exposing previously unknown patterns in their databases, which were used to improve profits, enhance customer services, and ultimately achieve a competitive advantage. In this paper, we present our unique data mining system DM-II (Data Mining Integration and Interestingness). DM-II is a PC-based system working in Windows 95/98/NT environment. Apart from the normal components of a data mining system, DM-II has a number of unique and advanced sub-systems. These sub-systems have been applied in many real-life applications, including education applications, insurance application, accident application, disease application, drug screening application, image classification, etc. Here, we focus on discussing the following sub-systems: CBA: CBA originally stands for Classification-Based on Associations [lo]. It has now been extended with a number of other advanced features. In all, CBA has the following capabilities: Building classifiers using association rules: Traditionally, association rule mining and classifier building are regarded as two different data mining tasks. CBA unifies the two tasks. It is able to ,generate association rules and builds a classifier using a special subset of the association rules. What is significant is that CBA, in general, produces more accurate classifiers compared to the state-ofthe-art classification system C4.5 [ 161. It also helps to solve some outstanding problems with the existing classification systems. Pruning and summarizing the discovered associations: One major problem with association

Discover the next generation of data-mining tools and technologyThis book brings together an international team of eighty experts to present readers with the next generation of data-mining applications. Unlike other publications that take a strictly academic and theoretical approach, this book features authors who have successfully developed data-mining solutions for a variety of customer types. Presenting their state-of-the-art methodologies and techniques, the authors show readers how they can analyze enormous quantities of data and make new discoveries by connecting key pieces of data that may be spread across several different databases and file servers. The latest data-mining techniques that will revolutionize research across a wide variety of fields including business, science, healthcare, and industry are all presented. Organized by application, the twenty-five chapters cover applications in:. Industry and business . Science and engineering . Bioinformatics and biotechnology . Medicine and pharmaceuticals. Web and text-mining . Security. New trends in data-mining technology. And much more . . .Readers from a variety of disciplines will learn how the next generation of data-mining applications can radically enhance their ability to analyze data and open the doors to new opportunities. Readers will discover:. New data-mining tools to automate the evaluation and qualification of sales opportunities. The latest tools needed for gene mapping and proteomic data analysis. Sophisticated techniques that can be engaged in crime fighting and preventionWith its coverage of the most advanced applications, Next Generation of Data-Mining Applications is essential reading for all researchers working in data mining or who are tasked with making sense of an ever-growing quantity of data. The publication also serves as an excellent textbook for upper-level undergraduate and graduate courses in computer science, information management, and statistics

SufRec, an algorithm for mining association rules: Recursivity and task parallelism

Data Mining: Mining Frequent Patterns, Associations Rules, and Correlations