An Efficient Algorithm for Incremental Mining of Association Rules

At present, the existing incremental mining algorithms of sequential patterns can not make full use of the mining information in original database, the updated database is scanned many times and the projected databases for frequent items are constructed. In this paper, we propose an incremental mining algorithm of sequential patterns based on Sequence tree, called ISPBS. ISPBS uses the sequence tree (Stree) to store the mining information in the original database. Sequence tree is a novel data structure, it is similar in structure to the prefix tree. But the sequence tree stores all the sequences in the original database. When the database is updated, ISPBS only deals with the incremental sequences and adds the sequences to the sequence tree without constructing the projected database. It can find all the sequential patterns in the updated database by scanning the sequence tree. Experiments show that ISPBS outperforms PrefixSpan and IncSpan in time cost.

Data mining is aimed to extract hidden acknowledge from large dataset, in order to exploit it for predicting future trends and make decisions. Extracting meaningful and useful candidate optimally is handled by several algorithms, mainly those based on exploring incoming data, which can lose information. To address this issue, this paper proposes an algorithm named Incremental Apriori (IncA) for discovering frequent itemsets in transaction databases, which is in fact a variant of the well-known Apriori algorithm. In IncA, we introduce a notion of promising items generated from the original database, an incremental technique applied on incremental database and a health check process to ensure candidate generation completeness. On the theoretical side, our algorithm exhibits the best computational complexity compared to the recent state-of-the-art algorithms. On the other hand, we tested the proposed approach on large synthetic databases. The obtained results prove that IncA reduces the running time as well as the search space and also show that our algorithm performs better than the Apriori algorithm.

Recently, HIC has formulated its strategic direction for the 21st century. In the new direction, HIC will not only serve as a payment agency but also become a health information provider to the public. Following the new strategic direction, HIC has set up a new Information Management Division to fulfil its goals. The research section within the new division has entered a new research area for data mining applications to the management of diseases. This is an attempt to apply data mining techniques to analyze the medical service transaction data for disease management rather than the data used in the well-studied applications to medical diagnosis. The disease management project undergoing at The Health Insurance Commission of Australia is aimed at identifying good or not so good medical service practice from a large transaction database. In this paper, we present a case study on our application to diabetes management. In this case study, we selected 7443 diabetes patients living in Fairfeld and extracted their Medicare service data in 1998 from the Medicare transactional database item sequences. After selecting the 26 most frequently used items, we further transferred the patients’ item sequences into a set of vectors, each components of which represent a group of services. The choose four service groups which are identified by the Clinic Advisory Group (CAG) of Diabetes to be the most essential tests for diabetic patients.

Sequential pattern is an important research topic in data mining and knowledge discovery. Sequential pattern is traditionally formed as (A, B) where A and B are frequent sequence in a transaction database. We extend this definition to include sequential patterns of forms (A, notB), (notA, B) and (notA, notB), which present negative sequential patterns among sequences. We call patterns of the form (A, B) positive sequential patterns, and patterns of the other forms negative patterns. Negative sequential patterns can also provide very useful insight view into the data set although they are different from positive ones. We put forward a discovery algorithm for mining negative sequential patterns from large transaction database in this paper.

Problem of finding frequent patterns has long been studied because it is very essential to data mining tasks such as association rule analysis, clustering, and classification analysis. Privacy preserving data mining is another important issue for this domain since most users do not want their private information to leak out. In this paper, we proposed an efficient approach for mining maximal frequent patterns from a large transactional database with privacy preserving capability. As for privacy preserving, we utilized prime number based data transformation method. We also developed a noble algorithm for mining maximal frequent patterns based on lattice structure. Extensive performance analysis shows the effectiveness of our approach.

One of the most important data mining problems is mining association rules. In this paper we consider discovering association rules from large transaction databases. The problem of discovering association rules can be decomposed into two sub-problems: find large itemsets and generate association rules from large itemsets. The second sub-problem is easier one and the complexity of discovering association rules is determined by complexity of discovering large itemsets. In this paper, we suggest Apriori-based algorithm for discovering large itemsets. Actually, we suggest a new procedure for large itemsets generation which is more efficient than the appropriate procedure of the original Apriori algorithm. For its implementation, we suggest a modified sort-merge-join algorithm, which is more efficient than nested-loop-join algorithm, which is suggested in the original Apriori algorithm. Besides, we propose a way in which Apriori Multiple finishes in just two iterations.

Using physiologically based pharmacokinetic modelling for in silico–in vitro–in vivo extrapolation to predict chemical exposure

Web mining is one of the mining technologies, which applies data mining techniques in large amount of web log data. Web navigational mining discovers users’ access patterns from web logs. This information can be used to identify the behavior of the web user. However, the web data will grow rapidly in the short time, and some of the web data may be antiquated. The user behavior may be changed when the new web data is inserted into and the old web data is deleted from web logs. Therefore, the user behavior must be re-discovered from the updated web logs. However, it is very time-consuming to re-find the users’ access patterns. Hence, many researchers pay attention to the incremental mining, which utilizes the previous mining results and finds new patterns just from the inserted or deleted part of the web logs such that the mining time can be reduced. The present paper proposes an efficient incremental web navigational mining algorithm for discovering web navigational patterns when the user sequences are inserted into and deleted from original database. It avoids re-finding the original web navigational patterns and re-counting the original candidate sequences. It uses lattice structure to keep the previous mining results such that just new candidate sequences need to be computed. Hence, the web navigational patterns can be obtained rapidly when the navigational sequence database is updated. Besides, maximal web navigational patterns can also be obtained easily by traversing the lattice structure. The experimental results show that the present algorithm is more efficient than the other approaches.

Association rule is an important research topic in data mining and knowledge discovery. Traditional algorithms for mining association rules are built on the binary attributes databases, which has three limitations. Firstly, it can not concern quantitative attributes; secondly, it treats each item with the same significance although different item may have different significance; thirdly, only the direct association rules are discovered. Mining fuzzy association rules has been proposed to address the first limitation. In this paper, we put forward an idea for mining indirect weighted association rules to resolve the other two limitations, and a discovery algorithm for mining both direct and indirect weighted fuzzy association rules by integrating these three extensions.

Association rules mining is an important research topic in data mining and knowledge discovery. Traditional algorithms for mining association rules are built on the binary attributes databases, which has three limitations. Firstly, it cannot concern quantitative attributes; secondly, only the positive association rules are discovered; thirdly, it treat each item with the same significance although different item may have different significance. In this paper, we put forward a discovery algorithm for mining positive and negative fuzzy weighted association rules to resolve these three limitations.

Retraction Note to: An improved memory adaptive up-growth to mine high utility itemsets from large transaction databases

The computation of minimal rare itemsets is a well known task in data mining, with numerous applications, e.g., drugs effects analysis and network security, among others. This paper presents a novel approach to the computation of minimal rare itemsets. First, we introduce a generalization of the traditional minimal rare itemset model called k-minimal rare itemset. A k-minimal rare itemset is defined as an itemset that becomes frequent or rare based on the removal of at least k or at most (k − 1) items from it. We claim that our work is the first to propose this generalization in the field of data mining. We then present a SAT-based framework for efficiently discovering k-minimal rare itemsets from large transaction databases. Afterwards, by partitioning the k-minimal rare itemset mining problem into smaller sub-problems, we aim to make it more manageable and easier to solve. Finally, to evaluate the effectiveness and efficiency of our approach, we conduct extensive experimental analysis using various popular datasets. We compare our method with existing specialized algorithms and CP-based algorithms commonly used for this task.

Sequential pattern is an important research topic in data mining and knowledge discovery. Traditional algorithms for mining sequential patterns are built on the binary attributes databases, which has two limitations. First, it can not concern quantitative attributes; second, only direct sequential patterns are discovered. Mining fuzzy sequential patterns has been proposed to address the first limitation. In this paper, we put forward a discovery algorithm for mining indirect sequential patterns to resolve the second limitation, and a discovery algorithm for mining both direct and indirect fuzzy sequential patterns by combining these two approaches.

A fundamental problem in chain convenience store is selecting items with respect to notion of "cross-selling effect". Recent works have proven the problem is NP-hard. In this paper, genetic algorithm is applied to this problem. Based on the features of genetic algorithm, a method of defining imprecise fitness function is proposed which is suited for solving item selection with imprecision cross-selling effect. Quantitative analysis method on "cross-selling effect" is also presented in this paper, which employs the ideas of quantitative association rules, a recent data mining technique to discover quantitative affinities in large transaction databases.

Sequential pattern mining is an important mining task for discovering sequential patterns along with their insight relationships in many real-world applications. In practice, sequence databases are kept changing over the time along with their business. For some reasons, some sequences in the database are asked to be deleted from the database. In order to have a synchronization of discovered sequential patterns with the database from which they have been discovered, the sequential pattern mining task is re-considered with many challenges. As the number of deleted sequences is often smaller than the size of the entire database, re-mining from scratch the updated database might incur a high cost because sequential pattern mining is a computationally expensive task. In this paper, our work aims at an efficient incremental mining solution to the sequential pattern mining task with sequence deletions. Different from the existing works, we propose an expanded prefix tree by extending the existing prefix tree with additional structures for capturing more necessary information for the incremental mining process. Based on this tree, we propose an incremental sequential pattern mining algorithm, SPMD, for finding a complete set of sequential patterns with no re-scanning the original database, when a number of sequences in the database are deleted. Experimental results on the benchmark databases have confirmed that our SPMD algorithm outperforms the re-mining from scratch with the PrefixSpan algorithm with less running time.