Association Rule Mining with Differential Privacy

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.

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

Efficiently mining frequent itemsets and association rules on the encrypted outsourced data remains a great challenge for the time-consuming ciphertext computations. Nowadays, it has been not well addressed for privacy-preserving frequent itemsets and association rule mining schemes with mining efficiency, dataset, and query confidentiality simultaneously. In this paper, we investigate the study of privacy issues on frequent itemset mining and association rule mining on outsourced data in a two-cloud model, where the data are encrypted and outsourced by multiple owners holding different public keys. We develop several secure computation protocols based on additively homomorphic cryptosystem and additive secret sharing, which enable the clouds could securely mine the frequent itemsets and association rules. Furthermore, we also design two kinds of frequent itemset and association rule query service models, i.e., service customers query the cloud-mined results, and service customers query with their own decided threshold. The proposed scheme not only supports the mining process on the data encrypted by multiple public keys without compromising the security of the datasets, query data and query results, but also offline users. In addition, the experimental results show that our query scheme is much more efficient than the state-of-the-art work.

A provided data from a transactional database, has supported technical development which can automatically find product association or items saved in the database. This finding association rules between saved product in database known as mining association rules. There are so many theories and algorithm developed for conducting mining association rules. One of algorithm developed is Apriori algorithm. This method, has a main goal to find the maximum frequent itemset. Next, this frequent itemset will be generated into associative rules which are not shown before in the database, become valuable information for considering materials in the decision process. Apriori algorithm is a interpretation technique of mining association rules, will be implemented into a web based software. On the software test which use in some different data, it’s concluded that time for mining association rules depends on the presence of every item in every transaction, total of transaction, minimum support and minimum confidence. For smaller value of minimum support and minimum confidence that entered, program will generate more association rules, vice versa.

Data mining has a great role for making market revenue by its decisive quality. Frequent Itemsets Mining (FIM) is very focused and emerging field of research of data mining. .FIM is used in association rule mining and other data analytic tasks. Frequent itemsets are patterns those occur very frequently in dataset. This paper provides efficacious and in-depth study of FIM algorithms. A comparative study of existing FIM algorithms and their variations are analyzed to find the limitations of these algorithms. Apriori ,Fp Growth, Eclat are basic methodologies for Frequent itemset mining. But there is issue for efficiency and memory consumption. Initially FIM was started in 1990’s and first algorithm, Apriori has drawback of scanning database multiple time, so complexity and memory consumption was high. Then Fp-Growth algorithm required scanning of dataset twice to form Fp-tree so huge memory consumption was required to hold the fp tree for Frequent Itemset Mining. After it Parallel frequent mining (shared and non-shared memory) overcome the drawback of FP-Growth algorithms by balancing the workload of single node to achieve efficiency. Now in present era of big dataset with voluminous and growing data at every moment, more prominent, efficient and scalable algorithm is required as complexity of these algorithms exponentially increases with the growth of rapid data .So objective of the study is in-depth literature review of Frequent Itemsets Mining algorithms to develop an efficient FIM algorithm.

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.

Association Rule Mining (ARM) is a well recognized and interesting area of research in the field of Data Mining. In ARM, significant amount of research work has been reported. On the contrary, very less work has been reported for Negative Association Rule Mining (NARM). ARM concentrates only on positive rules while NARM explores negative rules. Also, most researchers have worked with discrete transaction dataset. So, in this paper, we propose a technique called Negative and Positive Fuzzy Association Rule Mining (NP-FARM), which mines both negative and positive association rules from a fuzzy transaction dataset. NP-FARM algorithm has been implemented and the experimental results determine the optimal minimum support threshold and optimal minimum confidence threshold for the given dataset. Also, the experimental results demonstrate that, as the size of the dataset increases, a negligible change in execution time is witnessed to mine the growing dataset. Keywords: Association Rule Mining. Fuzzy Transactions, Positive and Negative Item Sets

A novel software defect prediction approach via weighted classification based on association rule mining

Soft set based association rule mining

The study of Influencing factors of traffic accidents is an important research direction in the field of traffic safety. In this paper, the traffic accidents of Shanghai Expressway from April to June 2014 were excavated using association rule mining which generated lots of frequent item sets. The strong rules hidden in these frequent item sets often uncover the association between influencing factors of accidents, which can be used to reduce the occurrence of accidents by breaking them. The rules can also be used to probe usual scenes of accidents, and some corresponding security improvement measures can be taken to prevent the accidents, and ultimately improve the city's traffic safety level. General speaking, association rule mining can produce tons of weak rules, the study first designed a method to calculate minimal Support value of training parameters, and further put forward a way to extract strong rules automatically. The results of the experiments showed that these methods proposed in the paper are effective. Therefore, an automatic modeling algorithm using association rules was finally established to promote the effective application of association rule mining on intelligent transportation system.

In the recent years, the growth of online marketplace explodes and has become the new norm of shopping. Access to the internet allows consumers to visit local and global online marketplace. By using online shopping, consumers can view the latest products and comparing products' prices. Online marketplace offers flexibility and eases consumers in so many ways. On the contrary, it also has some limitations. Dynamic pricing allows sellers to enhance their marketing strategy by ensuring price competitive with other sellers. The frequent occurrence of real-time price changing limits the user to get the best deal. This paper discusses the analysis of price changes using Association Rule Mining. Price-ChARM finds a frequent pattern of price changes in corresponding to different timelines. On the online marketplace, due to multiple sellers can sell the same item with different offers such as delivery speed and dynamic pricing, the comparison can be quite tricky. We collected data from two well-known portals and we represent the dataset as set of prices for different products for the purpose of frequent itemset mining. We implement the Apriori algorithm and use a total amount of 3,960 records. We generate association rules from the frequent itemsets found in the records and visualize the confidence of 0.9 rules. The association rules represent the pattern of price changes in the portals. This study eases consumer shopping experiences by understanding the trend of price changes to provide a better decision in making a purchase. Â

This research demonstrates the application of association rule mining to spatio‐temporal data. Association rule mining seeks to discover associations among transactions encoded in a database. An association rule takes the form A→B where A (the antecedent) and B (the consequent) are sets of predicates. A spatio‐temporal association rule occurs when there is a spatio‐temporal relationship in the antecedent or consequent of the rule. As a case study, association rule mining is used to explore the spatial and temporal relationships among a set of variables that characterize socioeconomic and land cover change in the Denver, Colorado, USA region from 1970–1990. Geographic Information Systems (GIS)‐based data pre‐processing is used to integrate diverse data sets, extract spatio‐temporal relationships, classify numeric data into ordinal categories, and encode spatio‐temporal relationship data in tabular format for use by conventional (non‐spatio‐temporal) association rule mining software. Multiple level association rule mining is supported by the development of a hierarchical classification scheme (concept hierarchy) for each variable. Further research in spatio‐temporal association rule mining should address issues of data integration, data classification, the representation and calculation of spatial relationships, and strategies for finding ‘interesting’ rules.

Multi-dimensional semantic clustering of large databases for association rule mining

In order to solve these problems how to easily generate candidate frequent item sets and fast compute support of candidate item sets, an algorithm of association rules mining based on binary has been introduced. However, one presented binary mining algorithm is only suitable for mining some relative short frequent item sets since the way of generating candidate item sets is also similar to Apriori, another is only suitable for mining long frequent item sets, which generates candidate item sets by up-down search strategy. And so aiming to mining general frequent item sets, this paper proposes an algorithm of association rules double search mining based on binary, which is different from tradition association rules mining algorithm based on double search strategy. The algorithm doesn’t use combination of set theory to generate candidate item sets but binary logic operation that is also used to compute support of candidate item sets, which can use character digital to reduce the number of scanned transaction. The algorithm gets rid of shortage about some presented algorithms based on binary. The experiment based on above three algorithms indicates that the efficiency about double search strategy is fast and efficient when mining general frequent item sets which aren’t confined.

Personal data privacy is a great concern by governments across the world as citizens generate huge amount of data continuously and industries using this for betterment of user centric services. There must be a reasonable balance between data privacy and utility of data. Differential privacy is a promise by data collector to the customer’s personal privacy. Centralised Differential Privacy (CDP) is performing output perturbation of user’s data by applying required privacy budget. This promises the inclusion or exclusion of individual’s data in data set not going to create significant change for a statistical query output and it offers -Differential privacy guarantee. CDP is holding a strong belief on trusted data collector and applying global sensitivity of the data. Local Differential Privacy (LDP) helps user to locally perturb his data and there by guaranteeing privacy even with untrusted data collector. Many differential privacy algorithms handles parameters like privacy budget, sensitivity and data utility in different ways and mostly trying to keep trade-off between privacy and utility of data. This paper evaluates differential privacy algorithms in regard to the privacy support it offers according to the sensitivity of the data. Generalized application of privacy budget is found ineffective in comparison to the sensitivity based usage of privacy budget.