Distributed Mining of Maximal Frequent Itemsets on a Data Grid System

In this paper, we propose a new algorithm, named Grid-based Distributed Max-Miner (GridDMM), for mining maximal frequent itemsets from databases on a Data Grid. A frequent itemset is maximal if none of its supersets is frequent. GridDMM is specifically suitable for use in Grid environments due to low communication and synchronization overhead. GridDMM consists of a local mining phase and a global mining phase. During the local mining phase, each node mines the local database to discover the local maximal frequent itemsets, then they form a set of maximal candidate itemsets for the top-down search in the subsequent global mining phase. A new prefix-tree data structure is developed to facilitate the storage and counting of the global candidate itemsets of different sizes. We built a Data Grid system on a cluster of workstations using the open-source Globus Toolkit, and evaluated the GridDMM algorithm in terms of performance, scalability, and the overhead of communication and synchronization. GridDMM demonstrates better performance than other sequential and parallel algorithms, and its performance is scalable in terms of the database size and the number of nodes. This research was supported in part by LexisNexis, NCR and AFRL/Wright Brothers Institute (WBI).     

Efficient mining of maximal frequent itemsets from databases on a cluster of workstations

In this paper, we propose two parallel algorithms for mining maximal frequent itemsets from databases. A frequent itemset is maximal if none of its supersets is frequent. One parallel algorithm is named distributed max-miner (DMM), and it requires very low communication and synchronization overhead in distributed computing systems. DMM has the local mining phase and the global mining phase. During the local mining phase, each node mines the local database to discover the local maximal frequent itemsets, then they form a set of maximal candidate itemsets for the top-down search in the subsequent global mining phase. A new prefix tree data structure is developed to facilitate the storage and counting of the global candidate itemsets of different sizes. This global mining phase using the prefix tree can work with any local mining algorithm. Another parallel algorithm, named parallel max-miner (PMM), is a parallel version of the sequential max-miner algorithm (Proc of ACM SIGMOD Int Conf on Management of Data, 1998, pp 85–93). Most of existing mining algorithms discover the frequent k-itemsets on the kth pass over the databases, and then generate the candidate (k + 1)-itemsets for the next pass. Compared to those level-wise algorithms, PMM looks ahead at each pass and prunes more candidate itemsets by checking the frequencies of their supersets. Both DMM and PMM were implemented on a cluster of workstations, and their performance was evaluated for various cases. They demonstrate very good performance and scalability even when there are large maximal frequent itemsets (i.e., long patterns) in databases.

快速挖掘频繁项集的并行算法

传统的挖掘频繁项集的并行算法存在数据偏移、通信量大、同步次数较多和扫描数据库次数较多等问题。针对这些问题,提出了一种快速挖掘频繁项集的并行算法(FPMFI)。FPMFI算法让各计算机节点独立地计算局部频繁项集,然后与中心节点交互实现数据汇总,最终获得全局频繁项集。理论分析和实验结果表明FPMFI算法是有效的。     

An Adaptive Algorithm for Mining Association Rules on Shared-Memory Parallel Machines

Mining association rules from large databases is very costly. We propose to develop parallel algorithms for this task on shared-memory multiprocessor (SMP). All proposed parallel algorithms for other paradigms follow the conventional level-wise approach: they need as many iterations as the length of the maximum large itemset. To make matter worse, they impose a synchronization in every iteration which would cause serious I/O contention on shared-memory parallel system. An adaptive asynchronous parallel mining algorithm APM has been proposed for SMP. All processors generate candidates dynamically and count itemset supports independently without synchronization. Two optimization techniques have been proposed for the reduction of database scanning and the number of candidates. The algorithm APM has been implemented on a Sun Enterprise 4000 shared-memory multiprocessor with 12 nodes. The experiments show that the optimizations have very good effects and APM has a substantial lead in performance over other proposed level-wise algorithms.     

Parallel Algorithms for Discovery of Association Rules

Discovery of association rules is an important data mining task. Several parallel and sequential algorithms have been proposed in the literature to solve this problem. Almost all of these algorithms make repeated passes over the database to determine the set of frequent itemsets (a subset of database items), thus incurring high I/O overhead. In the parallel case, most algorithms perform a sum-reduction at the end of each pass to construct the global counts, also incurring high synchronization cost. In this paper we describe new parallel association mining algorithms. The algorithms use novel itemset clustering techniques to approximate the set of potentially maximal frequent itemsets. Once this set has been identified, the algorithms make use of efficient traversal techniques to generate the frequent itemsets contained in each cluster. We propose two clustering schemes based on equivalence classes and maximal hypergraph cliques, and study two lattice traversal techniques based on bottom-up and hybrid search. We use a vertical database layout to cluster related transactions together. The database is also selectively replicated so that the portion of the database needed for the computation of associations is local to each processor. After the initial set-up phase, the algorithms do not need any further communication or synchronization. The algorithms minimize I/O overheads by scanning the local database portion only twice. Once in the set-up phase, and once when processing the itemset clusters. Unlike previous parallel approaches, the algorithms use simple intersection operations to compute frequent itemsets and do not have to maintain or search complex hash structures. Our experimental testbed is a 32-processor DEC Alpha cluster inter-connected by the Memory Channel network. We present results on the performance of our algorithms on various databases, and compare it against a well known parallel algorithm. The best new algorithm outperforms it by an order of magnitude.     

自适应区间配置在关联规则并行采掘中的作用

现行的采掘关联规则的并行算法基于经典的层次算法.该方法在每一次重复扫描数据库时都需要一次同步,这种同步运算对于共享内存多处理器并行机来说极大地降低了采掘性能,这种低效主要源于对共享的I/O通道的竞争.该文提出了在共享内存多处理机上采掘关联规则的异步算法APM.在APM中,所有参与计算的处理器能独立地产生备选集和计算支持度.而且,APM所需的扫描数据库的次数比层次方法所需的更少.该文还提出了一种增强APM的技术,使得该算法的性能对于数据分布更具有鲁棒性.文中实现了APM的变种算法,还实现了Apriori的并行版本Count Distribution算法.在SGI Power Challenge SMP并行机上,进行了性能分析,结果表明所提出的异步算法APM具有更好的性能和可扩展性.     

分布环境中的并行频繁模式挖掘算法

频繁模式的并行挖掘算法是数据挖掘中重要的研究课题。目前已经提出的并行算法大多是基于Apriori或基于FP-tree。由于两者的固有局限性,而且在计算过程中需要多次同步,因而具有较低的性能。文章提出了一种基于分布数据库的并行挖掘算法。该算法尽可能地让每个处理器独立地挖掘,每个处理器基于前缀树采用深度优先搜索的策略挖掘局部频繁模式集,并通过相关性质尽量减少候选全局频繁模式的规模,减少网络的通信量和同步次数以提高挖掘效率。     

Efficient Algorithms for Mining and Incremental Update of Maximal Frequent Sequences

We study two problems: (1) mining frequent sequences from a transactional database, and (2) incremental update of frequent sequences when the underlying database changes over time. We review existing sequence mining algorithms including GSP, PrefixSpan, SPADE, and ISM. We point out the large memory requirement of Pref ixSpan, SPADE, and ISM, and evaluate the performance of GSP. We discuss the high I/O cost of GSP, particularly when the database contains long frequent sequences. To reduce the I/O requirement, we propose an algorithm MFS, which could be considered as a generalization of GSP. The general strategy of MFS is to first find an approximate solution to the set of frequent sequences and then perform successive refinement until the exact set of frequent sequences is obtained. We show that this successive refinement approach results in a significant improvement in I/O cost. We discuss how MFS can be applied to the incremental update problem. In particular, the result of a previous mining exercise can be used (by MFS) as a good initial approximate solution for the mining of an updated database. This results in an I/O efficient algorithm. To improve processing efficiency, we devise pruning techniques that, when coupled with GSP or MFS, result in algorithms that are both CPU and I/O efficient.This research is supported by Hong Kong Research Grants Council grant HKU 7040/02E.     

Mining constrained frequent itemsets from distributed uncertain data

Nowadays, high volumes of massive data can be generated from various sources (e.g., sensor data from environmental surveillance). Many existing distributed frequent itemset mining algorithms do not allow users to express the itemsets to be mined according to their intention via the use of constraints. Consequently, these unconstrained mining algorithms can yield numerous itemsets that are not interesting to users. Moreover, due to inherited measurement inaccuracies and/or network latencies, the data are often riddled with uncertainty. These call for both constrained mining and uncertain data mining. In this journal article, we propose a data-intensive computer system for tree-based mining of frequent itemsets that satisfy user-defined constraints from a distributed environment such as a wireless sensor network of uncertain data.     

Parallel Mining of Neuronal Spike Streams on Graphics Processing Units

Multi-electrode arrays (MEAs) provide dynamic and spatial perspectives into brain function by capturing the temporal behavior of spikes recorded from cultures and living tissue. Understanding the firing patterns of neurons implicit in these spike trains is crucial to gaining insight into cellular activity. We present a solution involving a massively parallel graphics processing unit (GPU) to mine spike train datasets. We focus on mining frequent episodes of firing patterns that capture coordinated events even in the presence of intervening background events. We present two algorithmic strategies—hybrid mining and two-pass elimination—to map the finite state machine-based counting algorithms onto GPUs. These strategies explore different computation-to-core mapping schemes and illustrate innovative parallel algorithm design patterns for temporal data mining. We also provide a multi-GPU mining framework, which exhibits additional performance enhancement. Together, these contributions move us towards a real-time solution to neuronal data mining.     

Mining top?k frequent patterns without minimum support threshold

Finding frequent patterns play an important role in mining association rules, sequences, episodes, Web log mining and many other interesting relationships among data. Frequent pattern mining methods often produce a huge number of frequent itemsets that is not feasible for effective usage. The number of highly correlated patterns is usually very small and may even be one. Most of the existing frequent pattern mining techniques often require the setting of many input parameters and may involve multiple passes over the database. Minimum support is the widely used parameter in frequent pattern mining to discover statistically significant patterns. Specifying appropriate minimum support is a challenging task for a data analyst as the choice of minimum support value is somewhat arbitrary. Generally, it is required to repeatedly execute an algorithm, heuristically tuning the value of minimum support over a wide range, until the desired result is obtained, certainly, a very time-consuming process. Setting up an inappropriate minimum support may also cause an algorithm to fail in finding the true patterns. We present a novel method to efficiently retrieve top few maximal frequent patterns in order of significance without use of the minimum support parameter. Instead, we are only required to specify a more human understandable parameter, namely the desired number itemsets k. Our technique requires only a single pass over the database and generation of length two itemsets. The association ratio graph is proposed as a compact structure containing concise information, which is created in time quadratic to the size of the database. Algorithms are described for using this graph structure to discover top-most and top-k maximal frequent itemsets without minimum support threshold. To effectively achieve this, the method employs construction of an all path source-to-destination tree to discover all maximal cycles in the graph. The results can be ranked in decreasing order of significance. Results are presented demonstrating the performance advantages to be gained from the use of this approach.     

基于VOPP并行编程环境的最大频繁项集生成方法*

频繁项集挖掘是关联规则挖掘的核心内容,提出了一种挖掘最大频繁项集的并行算法CDTR。它对CD (counting distribution)算法进行了改进,根据一种新的分布式共享内存环境下面向视图并行编程思想,将数据库划分成视图。为了实现动态任务分配,对数据库进行了预处理。实验结果显示CDTR能够高效地生成最大频繁项集,大大提高了分布式共享内存系统的效率。     

无序嵌入式频繁子树挖掘算法

频繁模式挖掘的研究对象包括事务、序列、树和图。该文提出用模式增长方法在无序树构成的森林中挖掘嵌入频繁子树。利用规范化方法实现用唯一的形式表现无序树,根据待增长模式的拓扑结构确定其增长点并构造相应的投影库,将挖掘频繁子树模式问题转化为在各个投影库中寻找频繁节点的问题。     

Sequential Pattern Mining in Multi-Databases via Multiple Alignment

To efficiently find global patterns from a multi-database, information in each local database must first be mined and summarized at the local level. Then only the summarized information is forwarded to the global mining process. However, conventional sequential pattern mining methods based on support cannot summarize the local information and is ineffective for global pattern mining from multiple data sources. In this paper, we present an alternative local mining approach for finding sequential patterns in the local databases of a multi-database. We propose the theme of approximate sequential pattern mining roughly defined as identifying patterns approximately shared by many sequences. Approximate sequential patterns can effectively summerize and represent the local databases by identifying the underlying trends in the data. We present a novel algorithm, ApproxMAP, to mine approximate sequential patterns, called consensus patterns, from large sequence databases in two steps. First, sequences are clustered by similarity. Then, consensus patterns are mined directly from each cluster through multiple alignment. We conduct an extensive and systematic performance study over synthetic and real data. The results demonstrate that ApproxMAP is effective and scalable in mining large sequences databases with long patterns. Hence, ApproxMAP can efficiently summarize a local database and reduce the cost for global mining. Furthremore, we present an elegant and uniform model to identify both high vote sequential patterns and exceptional sequential patterns from the collection of these consensus patterns from each local databases.     

并行动态位向量频繁闭合序列模式挖掘算法

针对在时间和空间上都具有高计算成本的长序列数据库,一个更有效和更紧凑且可以完全提取信息的挖掘模式是当前的研究热点。提出一种并行动态位向量频繁闭合序列模式的挖掘算法（PDBV FCSP）,该算法采用多核处理器架构和DBV数据结构相结合的方式,有效加快了序列数据库的处理速度,并对搜索空间进行划分,尽早执行预处理序列的闭合检查,减少了所需的存储空间和挖掘频繁闭合序列模式的执行时间,克服了现有并行挖掘算法通信开销、同步和数据复制等问题。利用重新分配工作的动态负载平衡机制,解决处理器之间的负载均衡问题,最大限度地减少了CPU空闲时间。对DBV VDF算法和PDBV FCSP（2 4核）算法进行仿真比较,结果表明,PDBV FCSP算法在运行时间、内存使用和可伸缩性等方面都有较优的性能提升,且当内核数增加时,性能更优。     

标签集约束近似频繁模式的并行挖掘

近似频繁模式衍生于频繁模式,综合了频繁项集与频繁子图的特点。针对该模式的研究集中在无标签图上,其应用场景主要为社交网络、语义网络、智能电网等。近似频繁模式挖掘过程同时涉及频繁项集挖掘和频繁子图挖掘,因此已有的处理频繁模式挖掘算法无法较好地解决近似频繁模式挖掘问题。基于近似频繁模式结构,将其拓展到带标签图中,引入标签集约束,并设计标签集约束近似频繁模式挖掘算法LCPP（Label-Constraint Proximity Pattern）,该算法并行部署在MapReduce计算模型中,弥补了开源pFP算法处理大规模数据时效率不高的缺点。实验结果验证了该算法的有效性和可扩展性,表明了LCPP算法是pFP算法的极佳补充。     

MAXSeq: 一个新的最大频繁序列挖掘算法

最大频繁序列发现是数据挖掘中的一个重要分支．本文提出一种发现最大频繁序列集的算法MAXSeq，该算法通过对潜在的最大频繁序列进行选择性的扩展，直接判断其是否为最大序列，无须对候选最大序列进行维护，从而显著减小了存储开销．同时，优化策略的恰当运用对降低CPU时间起着至关重要的作用．     

一种有效的并行频繁项集挖掘算法

传统的挖掘频繁项集的并行算法存在各节点间负载不均衡、同步开销过大、通信量大等问题。针对这些问题,提出了一种多次传送重新分配数据的并行算法(MRPD)。MRPD算法在第l步时将数据库重新划分成若干组,并根据各节点的需要多次传送分组;各节点获得完整分组后异步地计算频繁项集;所有节点计算完成后,得到全部频繁项集。理论分析和实验结果表明MRPD算法是有效的。     

Novel parallel method for association rule mining on multi-core shared memory systems

Association rule mining (ARM) is an important task in data mining with many practical applications. Current methods for association rule mining have shown unstable performance for different database types and under-utilize the benefits of multi-core shared memory machines. In this paper, we address these issues by presenting a novel parallel method for finding frequent patterns, the most computational intensive phase of ARM. Our proposed method, named ShaFEM, combines two mining strategies and applies the most appropriate one to each data subset of the database to efficiently adapt to the data characteristics and run fast on both sparse and dense databases. In addition, our newlock-free design minimizes the synchronization needs and maximizes the data independence to enhance the scalability. The new structure lends itself well to dynamic job scheduling resulting in a well-balanced load on the new multi-core shared memory architectures. We have evaluated ShaFEM on 12-core multi-socket servers and found that our method run up to 5.8 times faster and consumes memory up to 7.1 times less than the state-of-the-art parallel method. For some test cases, ShaFEM can save up to 4.9 days of execution time over the compared method.     

Effective database transformation and efficient support computation for mining sequential patterns

In this paper, we propose a novel algorithm for mining frequent sequences from transaction databases. The transactions of the same customers form a set of customer sequences. A sequence (an ordered list of itemsets) is frequent if the number of customer sequences containing it satisfies the user-specified threshold. The 1-sequence is a special type of sequences because it consists of only a single itemset instead of an ordered list, while the k-sequence is a sequence composed of k itemsets. Compared with the cost of mining frequent k-sequences (k ≥ 2), the cost of mining frequent 1-sequences is negligible. We adopt a two-phase architecture to find the two types of frequent sequences separately in order that the discovery of frequent k-sequences can be well designed and optimized. For efficient frequent k-sequence mining, every frequent 1-sequence is encoded as a unique symbol and the database is transformed into one constituted by the symbols. We find that it is unnecessary to encode all the frequent 1-seqences, and make full use of the discovered frequent 1-sequences to transform the database into one with a smaller size. For every k ≥ 2, the customer sequences in the transformed database are scanned to find all the frequent k-sequences. We devise the compact representation for a customer sequence and elaborate the method to enumerate all distinct subsequences from a customer sequence without redundant scans. The soundness of the proposed approach is verified and a number of experiments are performed. The results show that our approach outperforms the previous works in both scalability and execution time.