Sue-Chen Hsueh

Apriori-based frequent itemset mining algorithms on MapReduce

Many parallelization techniques have been proposed to enhance the performance of the Apriori-like frequent itemset mining algorithms. Characterized by both map and reduce functions, MapReduce has emerged and excels in the mining of datasets of terabyte scale or larger in either homogeneous or heterogeneous clusters. Minimizing the scheduling overhead of each map-reduce phase and maximizing the utilization of nodes in each phase are keys to successful MapReduce implementations. In this paper, we propose three algorithms, named SPC, FPC, and DPC, to investigate effective implementations of the Apriori algorithm in the MapReduce framework. DPC features in dynamically combining candidates of various lengths and outperforms both the straight-forward algorithm SPC and the fixed passes combined counting algorithm FPC. Extensive experimental results also show that all the three algorithms scale up linearly with respect to dataset sizes and cluster sizes.

High utility pattern mining using the maximal itemset property and lexicographic tree structures

The problem of high utility mining is discovering all of the high utility itemsets in a transactional database. Most algorithms find high utility itemsets in two steps. The first step identifies all of the potential itemsets. The second step then determines the high utility itemsets from the set of potential itemsets. The large number of potential itemsets in the first step is generally the mining bottleneck. If we can reduce the number of potential itemsets, the mining performance can be improved significantly. In this paper, we use a maximal itemset property and propose an algorithm called UMMI (high Utility Mining using the Maximal Itemset property) to significantly reduce the number of potential itemsets in the first step. In the second step, UMMI uses an effective lexicographic tree structure to determine all of the high utility itemsets. In general, UMMI outperforms all three of the previously used algorithms, including CTU-PRO, an optimized TWU-mining algorithm, and Two-Phase, in our experiments using synthetic datasets. On average, UMMI is 5, 3, and 7 times faster than CTU-PRO, TWU-mining, and Two-Phase, respectively. In a real data experiment, UMMI is 6 times faster than Two-Phase. The other two algorithms are not capable of completing the mining step in a reasonable amount of time. UMMI uses an approximately fixed amount of memory, which is generally less than the other algorithms for each mining. The experimental results show that the proposed algorithm can mine the high utility itemsets efficiently. In addition, UMMI is linearly scalable with respect to the number of transactions.

Secure cloud storage for convenient data archive of smart phones

The importance of the data stored in the smart phones is increased as more applications are deployed and executed. Once the smart phone is damaged or lost, the valuable information treasured in the device is lost altogether. If cloud storage can be integrated with cloud services for periodical data backup of a mobile client, the risk of data lost can be minimized. However, the important data might be uncovered by a malicious third party during retrieval or transmission of information using wireless cloud storage without proper authentication and protection. Therefore, in this paper, we design an archive mechanism that integrates cloud storage, hybrid cryptography, and digital signatures to provide security requirements for data storage of mobile phones. Our mechanism not only can avoid malicious attackers from illegal access but also can share desired information with targeted friends by distinct access rights.

Fast discovery of sequential patterns in large databases using effective time-indexing

Sequential pattern mining algorithms can often produce more accurate results if they work with specific constraints in addition to the support threshold. Many systems implement time-independent constraints by selecting qualified patterns. This selection cannot implement time-dependent constraints, because the support computation process must validate the time attributes of every data sequence during mining. Therefore, we propose a memory time-indexing approach, called METISP, to discover sequential patterns with time constraints including minimum-gap, maximum-gap, exact-gap, sliding window, and duration constraints. METISP scans the database into memory and constructs time-index sets for effective processing. METISP uses index sets and a pattern-growth strategy to mine patterns without generating any candidates or sub-databases. The index sets narrow down the search space to the sets of designated in-memory data sequences, and speed up the counting of potential items within the indicated ranges. Our comprehensive experiments show that METISP has better efficiency, even with low support and large databases, than the well-known GSP and DELISP algorithms. METISP scales up linearly with respect to database size.

Mining Negative Sequential Patterns for E-commerce Recommendations

Sequential patterns in customer transactional databases are commonly mined for E-Commerce recommendations. In many practical applications, the absence of certain item-sets and sequences could have important implications. Mining frequent sequences comprising not only the occurrence but also the absence of certain sequences will increase the accuracy of product recommendations. A sequential pattern containing at least one absent item set is called a negative sequential pattern. In this paper, we formulate the problem of negative sequential pattern mining by introducing practical constraints and propose an algorithm called PNSP for the mining. The discovered patterns can then be more interesting and effective to use. The experimental results show that PNSP may discover negative sequential patterns for practical E-commerce applications.

Light-weight authentication and billing in mobile communications

We present an efficient scheme for mobile devices to conduct transactions demanding high security. The use of cryptographic mechanisms is minimized by applying block-ciphers and one-way hash primitives, which can be operated with low-end processors and very little working space. Undeniable billing can be achieved by applying one-way hash chains. Owing to one-way hash functions, the integrity of the masked transactions and the billing information can be verified, even when the masked information is transmitted in public. Complicated cryptographic mechanisms, also the regulations among roaming countries, thus can be eliminated so that a light-weight billing can be achieved.

Mining Closed Sequential Patterns with Time Constraints

The mining of closed sequential patterns has attracted researchers for its capability of using compact results to preserving the same expressive power as traditional mining. Many studies have shown that constraints are essential for applications of sequential patterns. However, time constraints have not been incorporated into closed sequence mining yet. Therefore, we propose an algorithm called CTSP for closed sequential pattern mining with time constraints. CTSP loads the database into memory and constructs time-indexes to facilitate both pattern mining and closure checking, within the pattern growth framework. The index sets are utilized to efficiently mine the patterns without generating any candidate or sub-database. The bidirectional closure checking strategy further speeds up the mining. The comprehensive experiments with both synthetic and real datasets show that CTSP efficiently mines closed sequential patterns satisfying the time constraints, and has good linear scalability with respect to the database size.

Incremental Discovery of Sequential Patterns Using a Backward Mining Approach

Common sequential pattern mining algorithms handle static databases. Once the data change, the previous mining result will be incorrect, and we need to restart the entire mining process for the new updated sequence database. Previous approaches, within either Apriori-based or projection-based framework, mine patterns in a forward manner. Considering the incremental characteristics of sequence-merging, we develop a novel technique, called backward mining, for efficient incremental pattern discovery. We propose an algorithm, called BSPinc, for incremental mining of sequential patterns using a backward mining strategy. Stable sequences, whose support counts remain unchanged in the updated database, are identified and eliminated from the support counting process. Candidate sequences generated using backward extensions can be mined recursively within the ever-shrinking space of the projected sequences. The experimental results show that BSPinc worked an average of 2.5 times faster than the well-known IncSpan algorithm and outperformed SPAM an average of 3 times faster.

Incremental update on probabilistic frequent itemsets in uncertain databases

Mining frequent itemsets in an uncertain database is a highly complicated problem. Most algorithms focus on improving the mining efficiency with the assumption that the database is static. Uncertain databases, however, are constantly updated with newly appended transactions like certain databases. Some patterns may become obsolete and new ones may emerge due to updates. Remining the whole uncertain database from scratch is very time-consuming owing to the frequentness probabilities computations. To tackle this maintenance problem, we propose an algorithm called p-FUP for efficient incremental update of patterns in an uncertain database. The p-FUP algorithm, inspired by a threshold-based PFI-testing technique and the FUP algorithm, uses approximations to incrementally update and discovers frequent itemsets in the uncertain database. Comprehensive experiments using both real and synthetic datasets show that p-FUP outperforms the re-mining based algorithm of 2.8 times faster in average, and has good linear scalability.

Mining Sequential Patterns for Image Classification in Ubiquitous Multimedia Systems

In this paper, a novel blind data hiding technique based on quantization is proposed. Quantization is performed in a special domain which converts one dimensional signal to a 2-D one named Point to Point Graph (PPG). Basis of the method is on the separation of this domain into two portions; while, only one portion is quantized. Furthermore, in the data extraction procedure, by using the unquantized portion and counting the number of samples, the embedded data is extracted. The performance of the proposed method is analytically investigated and verified by simulation with artificial Gaussian signals. Experimental simulations over audio signals show that the proposed method outperforms the recent quantization based watermarking approaches.