Traditional Association Rule Mining Research Articles

Multi-objective association rule mining with binary bat algorithm

Association rule mining meeting a variety of measures is regarded as a multi-objective optimization problem rather than a single objective optimization problem. The convergent speed of traditional multi-objective algorithms such as genetic algorithm is slow and the efficiency of these algorithms is low. Furthermore, the rules generated by traditional multi-objective algorithms are too large to be efficiently analyzed and explored in any further process. Bat algorithm is a new efficient global optimal algorithm whose convergence is superior to binary particle swarm optimization (BPSO) and genetic algorithm. This paper discusses the application of multi-objective bat algorithm to association rule mining. We propose multi-objective binary bat algorithm (MBBA) based on Pareto for association rule mining. This algorithm is independent of minimum support and minimum confidence. To evaluate the association rules mined by MBBA algorithm, we propose a new method to discover interesting association rules without favoring or excluding any measure. Compared with the single-objective BPSO, binary bat algorithm (BBA) and Apriori algorithm, the experimental results on six datasets show that the new algorithm is feasible and highly effective. It can make up the shortage of single objective algorithms and traditional association rule mining algorithms.

A Novel Mapreduce Lift Association Rule Mining Algorithm (Mrlar) for Big Data

Big Data mining is an analytic process used to discover the hidden knowledge and patterns from a massive, complex, and multi-dimensional dataset. Single-processor's memory and CPU resources are very limited, which makes the algorithm performance ineffective. Recently, there has been renewed interest in using association rule mining (ARM) in Big Data to uncover relationships between what seems to be unrelated. However, the traditional discovery ARM techniques are unable to handle this huge amount of data. Therefore, there is a vital need to scalable and parallel strategies for ARM based on Big Data approaches. This paper develops a novel MapReduce framework for an association rule algorithm based on Lift interestingness measurement (MRLAR) which can handle massive datasets with a large number of nodes. The experimental result shows the effi-ciency of the proposed algorithm to measure the correlations between itemsets through integrating the uses of MapReduce and LIM instead of depending on confidence.

Knowledge discovery of digital library subscription by RFC itemsets

Purpose The paper aims to understand the book subscription characteristics of the students at each college and help the library administrators to conduct efficient library management plans for books in the library. Unlike the traditional association rule mining (ARM) techniques which mine patterns from a single data set, this paper proposes a model, recency-frequency-college (RFC) model, to analyse book subscription characteristics of library users and then discovers interesting association rules from equivalence-class RFC segments. Design/methodology/approach A framework which integrates the RFC model and ARM technique is proposed to analyse book subscription characteristics of library users. First, the author applies the RFC model to determine library users’ RFC values. After that, the author clusters library users’ transactions into several RFC segments by their RFC values. Finally, the author discovers RFC association rules and analyses book subscription characteristics of RFC segments (library users). Findings The paper provides experimental results from the survey data. It shows that the precision of the frequent itemsets discovered by the proposed RFC model outperforms the traditional approach in predicting library user subscription itemsets in the following time periods. Besides, the proposed approach can discover interesting and valuable patterns from library book circulation transactions. Research limitations/implications Because RFC thresholds were assigned based on expert opinion in this paper, it is an acquisition bottleneck. Therefore, researchers are encouraged to automatically infer the RFC thresholds from the library book circulation transactions. Practical implications The paper includes implications for the library administrators in conducting library book management plans for different library users. Originality/value This paper proposes a model, the RFC model, to analyse book subscription characteristics of library users.

An efficient method for discovery of large item sets

In today's emerging field of descriptive data mining, association rule mining (ARM) has been proven helpful to describe essential characteristics of data from large databases. Mining frequent item sets is the fundamental task of ARM. Apriori, the most influential traditional ARM algorithm, adopts iterative search strategy for frequent item set generation. But, multiple scans of database, candidate item set generation and large load of system's I/O are major abuses which degrade the mining performance of it. Therefore, we proposed a new method for mining frequent item sets which overcomes these shortcomings. It judges the importance of occurrence of an item set by counting present and absent count of an individual item. Performance evaluation with Apriori algorithm shows that proposed method is more efficient as it finds fewer items in frequent item set in 50% less time without backtracking. It also reduces system I/O load by scanning the database only once.

Genetic Based Hesitation Information Mining for Profitability Management

Traditional Association Rule Mining has been extensively used to discover interesting rules or relationships between items in large databases but it has limitations that it solely deals with the items or products that are sold but avoids the items that are nearly sold. These nearly sold things carry hesitation data since customers are indecisive to shop for them. In this paper, with the help of vague set theory, we describe that item’s hesitation information is precious knowledge for the design of profitable selling strategies. This work proposed Genetic Algorithm based on evolution principles that has found its strong base in mining or maximize the rules for the items that customers mostly hesitate to purchase or has a high percentage of hesitation because of some reasons like price of an item, quality of an item, etc. Fitness function, crossover, and mutation are the main parameters involved in Genetic Algorithm which we used in our work. This work describes that if the reason of giving up the items is identified and resolved, we can easily remove this hesitation status of a customer and considering newly evolved rules as the interesting ones for boosting the sales of the item.

Maintaining the discovered high-utility itemsets with transaction modification

Most approaches for discovering frequent itemsets derive association rules from a binary database. Profit, cost, and quantity are not considered in traditional association-rule mining. Utility mining was proposed to measure the utilities of purchase products to derive highutility itemsets (HUIs). Many algorithms have been proposed to efficiently find HUIs from a static database. In real-world applications, transactions are inserted, deleted, or modified in dynamic situations. Existing batch approaches have to re-process the updated database since previously discovered HUIs are not maintained. In this paper, a Fast UPdated (FUP) strategy with utility measure and a maintenance algorithm, called FUP-HUI-MOD, are developed to efficiently maintain and update discovered HUIs. When transactions are modified, the proposed algorithm partitions the transactions before and after the modification into two parts, creating four cases. Each case is maintained using a specific procedure to update the discovered HUIs. Based on the designed FUP-HUI-MOD algorithm, the original database is not required to be rescanned each time compared to the state-of-the-art high-utility itemset mining algorithms in batch mode. Experiments are conducted to show that the proposed algorithm outperforms batch algorithms in maintaining HUIs.

Optimum Frequent Pattern Approach for Efficient Incremental Mining on Large Databases using Map Reduce

Data mining defines hidden pattern in data sets and association between the patterns. In data mining, association rule mining is key technique for discovering useful patterns from large collection of data. Frequent itemset mining is a famous step of association rule mining. Frequent itemset mining is used to gather item sets after discovering association rules. Some limitations exist with the traditional association rule mining algorithms for large-scale data. As for FP-Growth algorithm, the success is limited by internal memory size because mining process is on the base of large tree-form data structure. A new traditional approach, FPgrowth technique is very efficient in large amount of data. FPGrowth algorithm constructs conditional frequent pattern tree and conditional pattern based from database which satisfies the minimum support. However, FP growth algorithm requires a tree storage structure, which results in high computation time. The proposed algorithm realizes to construct Optimum pattern Tree with the node as the data item of the transaction. This rare algorithm is implemented on Hadoop to reduce the computation cost. The Hadoop environment supports for handling the large data and process them in parallel manner for better performance. The optimal frequent pattern is obtained that satisfies the minimum support and confidence value.

An efficient approach for mining association rules from high utility itemsets

Traditional association rule mining based on the support–confidence framework provides the objective measure of the rules that are of interest to users. However, it does not reflect the semantic measure among the items. The semantic measure of an itemset is characterized with utility values that are typically associated with transaction items, where a user will be interested to an itemset only if it satisfies a given utility constraint. In this paper, we first define the problem of finding association rules using utility-confidence framework, which is a generalization of the amount-confidence measure. Using this semantic concept of rules, we then propose a compressed representation for association rules having minimal antecedent and maximal consequent. This representation is generated with the help of high utility closed itemsets (HUCI) and their generators. We propose the algorithms to generate the utility based non-redundant association rules and methods for reconstructing all association rules. Furthermore, we describe the algorithms which generate high utility itemsets (HUI) and high utility closed itemsets with their generators. These proposed algorithms are implemented using both synthetic and real datasets. The results demonstrate better efficiency and effectiveness of the proposed HUCI-Miner algorithm compared to other well-known existing algorithms. In addition, the experimental results show better quality in the compressed representation of the entire rule set under the considered framework.

Fuzzy association rule mining for fashion product development

Purpose – The emergence of the fast fashion trend has exerted a great pressure on fashion designers who are urged to consider customers’ preferences in their designs and develop new products in an efficient manner. The purpose of this paper is to develop a fuzzy association rule mining (FARM) approach for improving the efficiency and effectiveness of new product development (NPD) in fast fashion. Design/methodology/approach – The FARM identifies the hidden relationships between product styles and customer preferences. The knowledge discovered help the fashion industry design new products which are not only fashionable, but are also saleable in the market. Findings – To evaluate the proposed approach, a case study is conducted in a Hong Kong-based fashion company in which a real-set of data are tested to generate fuzzy association rules. The results reveal that the FARM approach can provide knowledge support to the fashion industry during NPD, shorten the NPD cycle time, and increase customer satisfaction. Originality/value – Compared with traditional association rule mining, the proposed FARM approach takes the fuzziness of data into consideration and the knowledge represented in the fuzzy rules is in a more human-understandable structure. It captures the voice of the customer into fashion product development and provides a specific solution to deal with the challenges brought by fast fashion. In addition, it helps increase the innovation and technological capability of the fashion industry.

The Research of Generation Algorithm of Frequent Itemsets in High-Dimensional Data

In the mining of association rules, the generation of frequent itemsets is a key factor that influence the efficiency and performance of the algorithm. With the increase of data dimension, it is obvious that the traditional association rules mining algorithm can’t meet the demand of high dimensional data mining. On the basis of Apriori algorithm, we put forward Split Mtrix _Apriori algorithm in this paper. By generating the Boolean matrix of the database, Split Mtrix _Apriori algorithm decreased the times of scanning database when generating the frequent itemsets. With adopting grouping processing strategy in the Boolean matrix, the algorithm can still keep high efficiency in dealing with high-dimensional data.So Split Mtrix _Apriori improved the efficiency of association rule mining significantly.

An Incremental High‐Utility Mining Algorithm with Transaction Insertion

Association-rule mining is commonly used to discover useful and meaningful patterns from a very large database. It only considers the occurrence frequencies of items to reveal the relationships among itemsets. Traditional association-rule mining is, however, not suitable in real-world applications since the purchased items from a customer may have various factors, such as profit or quantity. High-utility mining was designed to solve the limitations of association-rule mining by considering both the quantity and profit measures. Most algorithms of high-utility mining are designed to handle the static database. Fewer researches handle the dynamic high-utility mining with transaction insertion, thus requiring the computations of database rescan and combination explosion of pattern-growth mechanism. In this paper, an efficient incremental algorithm with transaction insertion is designed to reduce computations without candidate generation based on the utility-list structures. The enumeration tree and the relationships between 2-itemsets are also adopted in the proposed algorithm to speed up the computations. Several experiments are conducted to show the performance of the proposed algorithm in terms of runtime, memory consumption, and number of generated patterns.

Infrequent Weighted Item Set Mining in Complex Data Analysis

Infrequent Weighted Association Mining (IWAM) is one of the main areas in data mining for extracting the rare items in high dimensional datasets. Traditional Association rule mining algorithms produce large number of candidate sets along with the database scans. Due to large number of transactions and database size, traditional methods consume more time to find the relevant association rules with the specified threshold. Prior and post database scans are required an additional effort to validate the association rules. Most of the existing weighted models are implemented for mining frequent itemsets, but finding infrequent itemset mining are useful in many recent fields like web,medical,cloud,complex databases,protein sequence etc. In weighted infrequent association rule mining, each item in the transaction is assigned a weight in order to mine high utility infrequent itemsets. In this proposed work, weighted association rule mining algorithm is proposed to find infrequent itemsets using weighted threshold measures. Proposed approach gives better results on real-time datasets compare to existing weighted models.

Updating high-utility pattern trees with transaction modification

Traditional association-rule mining only concerns the occurrence frequencies of the items in a binary database. In real-world applications, customers may buy several copies of the purchased items. Other factors such as profit, quantity, or price should be concerned to measure the utilities of the purchased items. High-utility itemsets mining was thus proposed to consider the factors of quantity and profit. Two-phase model was the most commonly way to keep the transaction-weighted utilization downward closure property, thus reducing the numerous candidates in utility mining. Most methods for finding high-utility itemsets are used to handle a static database. In practical applications, transactions are changed whether insertion, deletion, or modification. Some itemsets may arise as the new high-utility itemsets or become invalid knowledge in the updated database. In this paper, a maintenance Fast Updated High Utility Pattern tree for transaction MODification (FUP-HUP-tree-MOD) algorithm is thus proposed to effective maintain and update the built HUP tree for mining high-utility itemsets in dynamic databases without candidate generation. Experiments are conducted to show better performance of the proposed algorithm compared to the two-phase algorithm and the HUP tree algorithm in batch mode.

Association Rule Mining Considering Local Frequent Patterns with Temporal Intervals

In traditional association rule mining algorithms, if the minimum support is set too high, many valuable rules will be lost. However, if the value is set too low, then numerous trivial rules will be generated. To overcome the difficulty of setting minimum support values, global and local patterns are mined herein. Owing to the temporal factor in association rule mining, an itemset may not occur frequently in the entire dataset (meaning that it is not a global p attern), but it may appear frequently over specific intervals (meaning that it is a local pattern). This paper proposed a temporal ass ociation rule mining algorithm for interval frequent-patterns, called GLFMiner, which automatically and efficiently generates all intervals w ithout prior domain knowledge in an efficient manner. GLFMiner considers not only global frequent-patterns, but also local frequent-patterns. Using the same value of minimum support, it can locate many valuable temporal rules without losing the rules that traditio nal algorithms may find. Experimental results reveal that our novel algorithm mines more temporal frequent-patterns than traditional association rule mining algorithms and is effective in real-world applications such as market basket analysis and intrusion detection systems.

Principal Association Mining: An efficient classification approach

Classification is one of the key tasks in business intelligence, decision science, and machine learning. Associative classification has aroused significant research interest in recent years due to its superior accuracy. Traditional association rule mining algorithms often yield many redundant and sometimes conflicting class association rules. This paper presents a new, efficient associative classification approach. This new approach produces a compact classifier with a small number of association rules, yet with good classification performance. This approach is based on a novel rule quality metric, named as Principality, which measures an association rule’s classification accuracy and coverage for a specific class. Heuristic methods utilizing the Principality metric are applied to rule pruning and associative classifier construction to produce a compact classifier. This Principal Association Mining (PAM) approach is confirmed to be effective at improving classification accuracy as well as decreasing classifier size by experiments conducted on 17 datasets.

Association Rule Generation using Modified Hashing Function

rule mining is one of the most interesting and challenging task in data mining process. There exists many association rule mining techniques, each having merits and demerits. The main problem that exists in many traditional association rule mining algorithms is that these algorithms need more than one database scan to generate association rules. As scanning the database is a costly operation, algorithms capable of generating association rules with only one scan is the need of the hour. In this paper, a novel algorithm for generating association rules is presented which uses hashing function. This algorithm scans the database only once by utilizing the latest version of priori algorithm, direct hashing algorithm and pruning process. The algorithm discovers set of association rules from frequent k-item sets by computing the frequency of each item set. Then pruning process is applied to minimize the number of item sets generated after scanning the size of the database. Experimental results show that our method is very effective in generating association rules without any collision, leading to very high data accuracy. KeywordsRule Mining, ARM, Hashing, Pruning, Basket

Mining useful association rules from questionnaire data

Traditional association rule mining algorithms often have difficulty handling questions that are implicitly related, producing rules that are very accurate but are so obvious as to be completely useless to researchers. This problem is compounded by the fact that standard objective measures for interestingness often capture the wrong information and are maximized by these obvious rules. We propose an enhancement to standard association rule mining that uses clustering to identify related questions to pre-prune rules involving similar questions, which are less likely to be subjectively interesting. This enhancement reduces the search space of rules, improving the algorithm's efficiency. In addition, the resulting rule list has a higher concentration of diverse rules, more likely to be useful to researchers. We demonstrate this improvement relative to existing algorithms on two real-world, public health questionnaires.

Semantic interestingness measures for discovering association rules in the skeletal dysplasia domain.

BackgroundLately, ontologies have become a fundamental building block in the process of formalising and storing complex biomedical information. With the currently existing wealth of formalised knowledge, the ability to discover implicit relationships between different ontological concepts becomes particularly important. One of the most widely used methods to achieve this is association rule mining. However, while previous research exists on applying traditional association rule mining on ontologies, no approach has, to date, exploited the advantages brought by using the structure of these ontologies in computing rule interestingness measures.ResultsWe introduce a method that combines concept similarity metrics, formulated using the intrinsic structure of a given ontology, with traditional interestingness measures to compute semantic interestingness measures in the process of association rule mining. We apply the method in our domain of interest – bone dysplasias – using the core ontologies characterising it and an annotated dataset of patient clinical summaries, with the goal of discovering implicit relationships between clinical features and disorders. Experimental results show that, using the above mentioned dataset and a voting strategy classification evaluation, the best scoring traditional interestingness measure achieves an accuracy of 57.33%, while the best scoring semantic interestingness measure achieves an accuracy of 64.38%, both at the recall cut-off point 5.ConclusionsSemantic interestingness measures outperform the traditional ones, and hence show that they are able to exploit the semantic similarities inherently present between ontological concepts. Nevertheless, this is dependent on the domain, and implicitly, on the semantic similarity metric chosen to model it.

Association Rules Mining on Heart Failure Differential Treatment Based on the Improved Firefly Algorithm

Research the heart failure medical cases of TCM (traditional Chinese medicine) to effectively mine the association rules of differential diagnosis and treatment. TCM medical cases are of vast amounts of data and strong relatedness, and a new and improved firefly algorithm based on the guide of normative knowledge has been proposed to overcome the shortcomings of traditional association rules mining algorithms with the handling of TCM medical cases data such as low efficiency, slow convergence rate and rules underreporting, etc. The algorithm sets the support degree threshold through the penalty function, adaptively adjusts the hunting zone by normative knowledge to improve the convergence rate and exploration ability of the algorithm; it uses the way of random disturbance to conduct disturbance operation so as to increase the population diversity and effectively avoid algorithm prematurity. Confirmatory experiment of TCM medical cases for the treatment of heart failure has been conducted, the experimental results show that this method has achieved a great improvement on individual diversity and the efficiency of effective rules extraction compared with traditional association rules mining algorithms, and the mining results are of a certain reference value for TCM clinical diagnosis and treatment of heart failure.

Web usage mining with evolutionary extraction of temporal fuzzy association rules

In Web usage mining, fuzzy association rules that have a temporal property can provide useful knowledge about when associations occur. However, there is a problem with traditional temporal fuzzy association rule mining algorithms. Some rules occur at the intersection of fuzzy sets’ boundaries where there is less support (lower membership), so the rules are lost. A genetic algorithm (GA)-based solution is described that uses the flexible nature of the 2-tuple linguistic representation to discover rules that occur at the intersection of fuzzy set boundaries. The GA-based approach is enhanced from previous work by including a graph representation and an improved fitness function. A comparison of the GA-based approach with a traditional approach on real-world Web log data discovered rules that were lost with the traditional approach. The GA-based approach is recommended as complementary to existing algorithms, because it discovers extra rules.