Frequent Association Rules Research Articles

Map Reduce Based Association Rule Mining from Big Data

In today’s world, the shopping is the largest fashionable trend where the transaction processing is meticulous to fetch the items from the shopping transaction history by using traditional Apriori algorithm. An Apriori algorithm is the one which is used for finding frequent pattern from the given dataset. The problem of Apriori is to find useful itemsets for business purpose was time consuming. To overcome this problem, we have proposed Map Reduce based Apriori algorithm which generates frequent itemset and association rules by using parallel computations to reduce computations. The Spark distributed systems along with data bricks technology have been used. The experimental result shows that have been reduced the time taken fetch the data from the database.

Utilizing graphics processing unit to accelerate drug-symptom association mining

A limited number of graphics processing unit algorithms exist for frequent itemset and association rule mining. This paper attempts to address that gap by introducing algorithms that lend themselves to massively parallel processing in a tool we call GPUMiner. The performance of GPUMiner will be contrasted against classic algorithms developed for a central processing unit type architecture. Multiple optimizations are adopted to improve efficiency in our design, including separate bitmaps for drugs and symptoms, parallel reduction for sum operation and a thread combination matrix that enables multiple-drug combinations to explored. Experiments, using the popular test dataset T40I10D100K.data, show that our GPUMiner is able to achieve a speedup of 13.7 in comparison to the existing implementation. In addition, we apply GPUMiner in discovering drug-symptom associations and report on some well-known symptoms associated with a single drug or a combination of multiple drugs.

Fast Dimensional Analysis for Root Cause Investigation in a Large-Scale Service Environment

Root cause analysis in a large-scale production environment is challenging due to the complexity of the services running across global data centers. Due to the distributed nature of a large-scale system, the various hardware, software, and tooling logs are often maintained separately, making it difficult to review the logs jointly for understanding production issues. Another challenge in reviewing the logs for identifying issues is the scale - there could easily be millions of entities, each described by hundreds of features. In this paper we present a fast dimensional analysis framework that automates the root cause analysis on structured logs with improved scalability. We first explore item-sets, i.e. combinations of feature values, that could identify groups of samples with sufficient support for the target failures using the Apriori algorithm and a subsequent improvement, FP-Growth. These algorithms were designed for frequent item-set mining and association rule learning over transactional databases. After applying them on structured logs, we select the item-sets that are most unique to the target failures based on lift. We propose pre-processing steps with the use of a large-scale real-time database and post-processing techniques and parallelism to further speed up the analysis and improve interpretability, and demonstrate that such optimization is necessary for handling large-scale production datasets. We have successfully rolled out this approach for root cause investigation purposes within Facebook's infrastructure. We also present the setup and results from multiple production use cases in this paper.

Determination of Temporal Association Rules Pattern Using Apriori Algorithm

A supermarket must have good business plan in order to meet customer desires. One way that can be done to meet customer desires is to find out the pattern of shopping purchases resulting from processing sales transaction data. Data processing produces information related to the function of the association between items of goods temporarily. Association rules functions in data mining.Association rule is one of the data mining techniques used to find patterns in combination of transaction data. Apriori algorithm can be used to find association rules. Apriori algorithm is used to find frequent itemset candidates who meet the support count. Frequent itemset that meets the support count is then processed using the temporal association rules method. The function of temporal association rules is as a time limitation in displaying the results of frequent itemsets and association rules. This study aims to produce rules from transaction data, apriori algorithm is used to form temporal association rules. The final results of this research are strong rules, they are rules that always appear in 3 years at certain time intervals with limitation on support and confidence, so that the rules can be used for business plan layout recommendations in Maharani Supermarket Demak.

Privacy-preserving frequent itemset mining in vertically partitioned database using symmetric homomorphic encryption scheme

Frequent itemset mining (FIM) and association rule mining (ARM) are two popular and widely used mining techniques for transactional databases with wide range of applications such as medical diagnosis, market basket analysis, protein sequences, census data, etc. In privacy-preserving frequent itemset mining (PPFIM) over vertically partitioned database, data owners wish to learn frequent itemsets from their concatenated dataset without disclosing their sensitive information. A collusion resistant scheme has been proposed using Du-Atallah's SBDP protocol for multiparty multi-vector scenario which works under two party two-vector case. Also, a scheme is designed by generalising the Du-Atallah's scheme for more than two parties. However it fails in privacy requirements when more than two parties/vectors are considered. In this paper, we give a critique on this approach and focus on efficient and more secure methods for privacy preserving FIM in vertically partitioned databases. To ensure privacy of raw data of different data owners having heterogeneous attributes, we proposed a secure-sum algorithm which uses symmetric homomorphic encryption scheme as a sub-part and a semihonest trusted third party (STTP) environment for calculating the support count of itemsets in privacy preserving manner.

Privacy-preserving frequent itemset mining in vertically partitioned database using symmetric homomorphic encryption scheme

Frequent itemset mining (FIM) and association rule mining (ARM) are two popular and widely used mining techniques for transactional databases with wide range of applications such as medical diagnosis, market basket analysis, protein sequences, census data, etc. In privacy-preserving frequent itemset mining (PPFIM) over vertically partitioned database, data owners wish to learn frequent itemsets from their concatenated dataset without disclosing their sensitive information. A collusion resistant scheme has been proposed using Du-Atallah's SBDP protocol for multiparty multi-vector scenario which works under two party two-vector case. Also, a scheme is designed by generalising the Du-Atallah's scheme for more than two parties. However it fails in privacy requirements when more than two parties/vectors are considered. In this paper, we give a critique on this approach and focus on efficient and more secure methods for privacy preserving FIM in vertically partitioned databases. To ensure privacy of raw data of different data owners having heterogeneous attributes, we proposed a secure-sum algorithm which uses symmetric homomorphic encryption scheme as a sub-part and a semihonest trusted third party (STTP) environment for calculating the support count of itemsets in privacy preserving manner.

An improved Frequent Pattern Mining in Sustainable Learning Practice using Generalized Association Rules

This research focuses on mining the frequent patterns occurred in the given Datasets by Generalization of Association Rules. Frequent pattern mining is a significant as well as interesting problem in the research filed of Data Mining. Building of frequent pattern tree (FP tree), frequent pattern growth (FP growth) and association rule generation are implemented to find the repeated patterns of data. FP tree Construction Algorithm is mainly used to compact a vast database into a extremely compressed, seems to very tiny data structure; hence eliminates the repeated scanning of database. The role of FP growth algorithm is to discover the frequent patterns with FP tree structure and construct the generalized association rules corresponding to the learning data. FP growth algorithm obtained best results as compared with conventional Apriori algorithm. This research provides some practical real time applications pertaining data mining techniques in the field of learning, education, marketing, finance and so on.

Various Research Opportunities in High Utility Itemset Mining

Pattern mining is a technique, which discovers interesting, hidden, unpredicted and useful patterns of data from the database. Most of the research work in pattern mining has been focused on the traditional way of Frequent Itemset Mining (FIM) and Association Rule Mining (ARM) for pattern-discovery. Patterns in frequent itemset mining are based on the occurrence frequency of items. Although frequent pattern mining is useful, the assumption that ‘frequent patterns are interesting,’ doesn’t hold for numerous applications. High Utility Itemset Mining (UIM) overcomes this limitation of frequent itemset mining. The aim of HUIM is to find the patterns based on a utility function where the utility can be measured in terms of revenue, profit, weight, frequency, interestingness or time spent on some webpage, etc. Mining patterns with high utility can be seen as a generalization of FIM where the transaction database is the input and every item is having a utility factor representing its importance and might have non-binary quantities in the transactions. This paper surveys various recent advances and research opportunities in the field of high utility itemset mining.

Assessment Method of the Economic Vulnerability of the Coastal Zone Based on the Analytic Hierarchy Process

Wang, B.; Nie, X.; Li, Z., and Fan, T., 2019. Assessment method of the economic vulnerability of the coastal zone based on the analytic hierarchy process. In: Guido-Aldana, P.A. and Mulahasan, S. (eds.), Advances in Water Resources and Exploration. Journal of Coastal Research, Special Issue No. 93, pp. 798–803. Coconut Creek (Florida), ISSN 0749-0208.It is an important strategic measure to carry out the scientific concept of development and build a harmonious society to make great efforts to build a coastal economic uplift zone and cultivate a new economic growth pole. A coastal economic vulnerability assessment method based on analytic hierarchy process (AHP) is proposed. And combined with statistical data for empirical analysis. The statistical information model of coastal economic vulnerability assessment is constructed. According to the mining results of coastal economic vulnerability assessment information, the frequent itemsets association rules of coastal economic vulnerability assessment data are reconstructed, the correlation dimension characteristics of coastal economic vulnerability information are extracted, and the regular characteristics of coastal economic vulnerability assessment cross-correlation information fusion are analyzed. Taking this as the constraint condition, the correlation information mining evaluation of coastal economic vulnerability assessment is carried out, and the statistical analysis and robustness test of coastal economic vulnerability are carried out by using statistical feature analysis method. The simulation structure shows that the model has good accuracy and high confidence in coastal economic vulnerability assessment, and improves the quality of coastal economic vulnerability assessment.

Identification of Risk Factors for Early Childhood Diseases Using Association Rules Algorithm with Feature Reduction

Abstract This paper introduces a technique that can efficiently identify symptoms and risk factors for early childhood diseases by using feature reduction, which was developed based on Principal Component Analysis (PCA) method. Previous research using Apriori algorithm for association rule mining only managed to get the frequent item sets, so it could only find the frequent association rules. Other studies used ARIMA algorithm and succeeded in obtaining the rare item sets and the rare association rules. The approach proposed in this study was to obtain all the complete sets including the frequent item sets and rare item sets with feature reduction. A series of experiments with several parameter values were extrapolated to analyze and compare the computing performance and rules produced by Apriori algorithm, ARIMA, and the proposed approach. The experimental results show that the proposed approach could yield more complete rules and better computing performance.

A data-driven approximate dynamic programming approach based on association rule learning: Spacecraft autonomy as a case study

Dynamic programming (DP) and Markov Decision Process (MDP) offer powerful tools for formulating, modeling, and solving decision making problems under uncertainty. In real-world applications, the applicability of DP is limited by severe scalability issues. These issues can be addressed by Approximate Dynamic Programming (ADP) techniques. ADP methods are based on the assumption of having either a proper estimation of the underlying state transition probability distributions or a simulation mechanism with the capability of generating samples according to such probability distributions. In this paper, we present a data-driven ADP-based approach, which can offer an alternative in case such assumption cannot be guaranteed. In particular, when varying the set-up of the MDP state transition probability matrix, different policies can be calculated through exact DP or ADP methods. Such policies are then processed by an Apriori-based algorithm to find frequent association rules within them. A pruning procedure is used to select the most suitable association rules, and finally an Association Classifier infers the optimal policy in all the possible circumstances. We show a detailed application of the proposed approach for the calculation of a proper mission operations plan for spacecrafts with a high level of on-board autonomy.

A Survey of Parallel Sequential Pattern Mining

With the growing popularity of shared resources, large volumes of complex data of different types are collected automatically. Traditional data mining algorithms generally have problems and challenges including huge memory cost, low processing speed, and inadequate hard disk space. As a fundamental task of data mining, sequential pattern mining (SPM) is used in a wide variety of real-life applications. However, it is more complex and challenging than other pattern mining tasks, i.e., frequent itemset mining and association rule mining, and also suffers from the above challenges when handling the large-scale data. To solve these problems, mining sequential patterns in a parallel or distributed computing environment has emerged as an important issue with many applications. In this article, an in-depth survey of the current status of parallel SPM (PSPM) is investigated and provided, including detailed categorization of traditional serial SPM approaches, and state-of-the art PSPM. We review the related work of PSPM in details including partition-based algorithms for PSPM, apriori-based PSPM, pattern-growth-based PSPM, and hybrid algorithms for PSPM, and provide deep description (i.e., characteristics, advantages, disadvantages, and summarization) of these parallel approaches of PSPM. Some advanced topics for PSPM, including parallel quantitative/weighted/utility SPM, PSPM from uncertain data and stream data, hardware acceleration for PSPM, are further reviewed in details. Besides, we review and provide some well-known open-source software of PSPM. Finally, we summarize some challenges and opportunities of PSPM in the big data era.

An Intelligent Decision in Smart Systems Using A Weighted Frequent Itemset Mining Algorithm

Intelligent decision is the key technology of smart systems. Data mining technology has been playing an increasingly important role in decision making activities. The introduction of weight makes the weighted frequent itemsets not satisfy the downward closure property any longer. As a result, the search space of frequent itemsets cannot be narrowed according to downward closure property which leads to a poor time efficiency. In this paper, the weight judgment downward closure property for weighted frequent itemsets and the existence property of weighted frequent subsets are introduced and proved first. The Fuzzy-based WARM satisfies the downward closure property and prunes the insignificant rules by assigning the weight to the itemset. This reduces the computation time and execution time. This paper presents an Enhanced Fuzzy-based Weighted AssociationRuleMining(E-FWARM) algorithm for efficient mining of the frequent itemsets. The pre-filtering method is applied to the input dataset to remove the item having low variance. Data discretization is performed and E-FWARM is applied for mining the frequent itemsets. The experimental results show that the proposed E-FWARM algorithm yields maximum frequent items, association rules, accuracy and minimum execution time than the existing algorithms.

Using Localized Features for Analyzing College Students’ Imagination

The analysis of imagination has become popular in recent years because imagination is one of the key components of creativity and innovation. For extracting students’ implicit degrees and thought processes of imagination, we use frequent pattern mining and association rule extraction to localize the features and explain the deep meanings of imagination in the study. By our observations, these two methods may sometimes explore meaningless frequent patterns and rules on a global sparse dataset. In order to eliminate such phenomena when mining with these two methods, we use a localized feature approach called forecast with clustering and integration (FCI) to improve the drawbacks of two methods on a sparse dataset. The approach consists of two strategies. One is clustering and the other is the prediction based on integration from (1) frequent patterns, (2) association rule pruning with correlation, and (3) forecast with linear regression. The former strategy can reduce the number of samples and highlight the features of imagination and the latter strategy can prune meaningless information and predict the trend of scores from imagination input data. Experimental results show both proposed approaches can localize special features, thereby providing supervisors with meaningful information about students’ degrees and thought processes of imagination.

Query Expansion in Information Retrieval using Frequent Pattern (FP) Growth Algorithm for Frequent Itemset Search and Association Rules Mining

Documents on the Internet have increased in number exponentially; this has resulted in users having difficulty finding documents or information needed. Special techniques are needed to retrieve documents that are relevant to user queries. One technique that can be used is Information Retrieval (IR). IR is the process of finding data (generally documents) in the form of text that matches the information needed from a collection of documents stored on a computer. Problems that often appear on IRs are incorrect user queries; this is caused by user limitations in representing their needs in the query. Researchers have proposed various solutions to overcome these limitations, one of which is to use the Expansion Query (QE). Various methods that have been applied to QE include Ontology, Latent Semantic Indexing (LSI), Local Co-Occurrence, Relevance Feedback, Concept Based, WordNet / Synonym Mapping. However, these methods still have limitations, one of them in terms of displaying the connection or relevance of the appearance of words or phrases in the document collection. To overcome this limitation, in this study we have proposed an approach to QE using the FP-Growth algorithm for the search for frequent itemset and Association Rules (AR) on QE. In this study, we applied the use of AR to QE to display the relevance of the appearance of a word or term with another word or term in the collection of documents, where the term produced is used to perform QE on user queries. The main contribution in this study is the use of Association rules with FP-Growth in the collection of documents to look for the connection of the emergence of words, which is then used to expand the original query of users on IR. For the evaluation of QE performance, we use recall, precision, and f-measure. Based on the research that has been done, it can be concluded that the use of AR on QE can improve the relevance of the documents produced. This is indicated by the average recall, precision, and f-measure values produced at 94.44%, 89.98%, and 92.07%. After comparing the IR process without QE with IR using QE, an increase in recall value was 25.65%, precision was 1.93%, and F-Measure was 15.78%.

Evaluation of relationships between onychomycosis and vascular diseases using sequential pattern mining

Onychomycosis (OM) is a common nail disease. Although controversial, vascular diseases are considered independent predictors of OM and vice versa. Sequential pattern mining (SPM) has not been previously used for statistical analysis in dermatology, but it is an efficient method for identifying frequent association rules in multiple sequential data sets. The aim of our study was to identify the relationship between OM and vascular diseases in the real world through a population-based study using SPM. We obtained population-based data recorded from 2002 to 2013 by the Health Insurance Research and Assessment Agency. Cases of vascular-related disease and OM were identified using the diagnostic codes of the International Classification of Diseases 10th Revision, version 2010. SPM measures were based on comorbidity and duration values. We estimated 3-year risk for progression from OM to vascular disease and vice versa using logistic regression. Patients with varicose veins and peripheral vascular disease had higher OM comorbidity (comorbidity: 1.26% and 0.69%, respectively) than did those with other vascular diseases. Patients diagnosed with varicose veins and peripheral vascular disease were diagnosed with OM after 25.50 and 55.10 days, respectively, which was a shorter duration than that observed for other diseases. Patients with OM were at higher risk for peripheral vascular disease (adjusted odds ratio (aOR) 1.199 [95% confidence interval (CI) 1.151–1.249]) and varicose veins (aOR 1.150 [95% CI 1.063–1.245]). Patients with peripheral vascular disease (aOR 1.128 [95% CI 1.081–1.177]) were at higher risk for OM, while patients with varicose veins had no significant risk for OM. Careful consideration of varicose veins or peripheral vascular disease is required for proper management of comorbidities in patients with OM.

Rare association rule mining from incremental databases

Rare association rule mining is an imperative field of data mining that attempts to identify rare correlations among the items in a database. Although numerous attempts pertaining to rare association rule mining can be found in the literature, there are still certain issues that need utmost attention. The most prominent one among them is the rare association rule mining from incremental databases. The existing rare association rule mining techniques are capable of operating only on static databases, assuming that the entire database to be operated on is available during the outset of the mining process. Inclusion of new records, however, may lead to the generation of some new interesting rules from the current set of data, invalidating the previously extracted significant rare association rules. Executing the entire mining process from scratch for the newly arrived set of data could be a tedious affair. With a view to resolve the issue of incremental rare association rule mining, this study presents a single-pass tree-based approach for extracting rare association rules when new data are inserted into the original database. The proposed approach is capable of generating the complete set of frequent and rare patterns without rescanning the updated database and reconstructing the entire tree structure when new transactions are added to the existent database. Experimental evaluation has been carried out on several benchmark real and synthetic datasets to analyze the efficiency of the proposed approach. Furthermore, to assess its applicability in real-world applications, experimental analysis has been performed on a real geological dataset where earthquake records are incrementally being added on an annual basis. Comparative performance analysis demonstrates the preeminence of proposed approach over existing frequent and rare association rule mining techniques.

Hiding sensitive itemsets without side effects

Data mining techniques are being used to discover useful patterns hidden in the data. However, these data mining techniques also extract sensitive information posing a threat to privacy. Frequent Itemset mining is a widely used data mining technique and a pre-processing step for Association Rule Mining. These frequent itemsets may contain sensitive itemsets which need to be hidden from adversaries. Traditional data sanitization techniques modify transactions in the database to hide sensitive itemsets which suffer from undesired side effects and information loss. In this paper, we propose a pattern sanitization approach to hide sensitive itemsets for privacy preserved pattern sharing. The transactional database is modeled as a set of lossless compact patterns using Closed Itemsets. The novelty of the proposed technique is in sanitizing the closed itemsets/patterns instead of transactions in the database. The proposed Recursive Pattern Sanitization (RPS) algorithm hides multiple sensitive itemsets irrespective of their size and support in single parse of the closed patterns. The patterns in the sanitized model retain the closeness property, and the model has inherent support for finding frequent itemsets and association rules reducing mining activity by the end user. Experimental results show that the proposed approach is effective in hiding sensitive itemsets without side effects and unexpected information loss compared to other well-known transaction modification based itemset hiding techniques.

Applying mutual information for discretization to support the discovery of rare-unusual association rule in cerebrovascular examination dataset

ABSTRACT In knowledge discovery studies, association rules mining has been extensively studied to discover hidden knowledge and relationships among set of items in a transactional dataset. Most research on association rule mining focuses on discovering frequent patterns based on the most frequent items occurring in the dataset. However, the process of extracting rare rules has received less attention. In medical dataset studies, the discovery of rare association rules (RARs) is more challenging, because it could likely be used to obtain more potentially rare and unusual knowledge for physicians, beside frequent association rules. Hence, the aim of this paper is to discover non-frequent or rare-unusual association rules (RUARs) from a stroke medical dataset to provide potential meaningful knowledge to the user domain. A discretization method needs to be performed as the data preprocessing step before generating rules. To the best of our knowledge, fewer studies have focused on the role of discretization results to support the extraction of a better amount and quality of RUARs, particularly for medical datasets. In addition, the extracted RUARs is expected to provide potential new unusual insights on stroke risk patterns. This paper applies mutual information measure to discretize a stroke examination dataset collected from a medical center in Taiwan. The interval merging method was proposed to simplify the discrete form and enrich the quality of generated rules. Towards the end, rare association rules, with relatively low support, were generated by employing the Apriori-Rare method accordingly. In addition, a filtering process was applied to the content of the rule itemsets to discover the expected set of RUARs for physicians. Furthermore, the extracted RUARs was analyzed based on the relative risk values toward the occurrence of stroke. Results indicated that the mutual information discretization outperformed the traditional discretization methods in terms of how the discretization scheme can support the extraction of RUARs with a better quantity and quality measurements for further analysis purpose in medical point of view. Moreover, the proposed method had a relatively higher number of RUARs. The knowledge of unusual rule patterns from rare association rules might provide potential new and unusual insights for medical pratitioners and increase the awareness of stroke examination results.

ARM–AMO: An efficient association rule mining algorithm based on animal migration optimization

Association rule mining (ARM) aims to find out association rules that satisfy predefined minimum support and confidence from a given database. However, in many cases ARM generates extremely large number of association rules, which are impossible for end users to comprehend or validate, thereby limiting the usefulness of data mining results. In this paper, we propose a new mining algorithm based on animal migration optimization (AMO), called ARM–AMO, to reduce the number of association rules. It is based on the idea that rules which are not of high support and unnecessary are deleted from the data. Firstly, Apriori algorithm is applied to generate frequent itemsets and association rules. Then, AMO is used to reduce the number of association rules with a new fitness function that incorporates frequent rules. It is observed from the experiments that, in comparison with the other relevant techniques, ARM–AMO greatly reduces the computational time for frequent item set generation, memory for association rule generation, and the number of rules generated.