Fast Algorithms for Frequent Itemset Mining from Uncertain Data

Leung, Carson K.; MacKinnon, Richard Kyle

doi:10.1109/icdm.2014.146

Cited by 51 publications

(11 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Experimental results show that, in terms of accuracy, MrCloud returned the same collection of frequent patterns as those returned by UF-growth [32], tube-growth [33] and BLIMP-growth [34]. Note that, in terms of flexibility, MrCloud is not confined to handling AM constraints with 100% selectivity.…”

Section: Evaluation Resultsmentioning

confidence: 86%

“…All experiments were run using either (i) a single machine with an Intel Core i7 4-core processor (1.73 GHz) and 8 GB of main memory running a 64-bit Windows 7 operating system; or (ii) the Amazon Elastic Compute Cloud (EC2) cluster-specifically, 11 High-Memory Extra Large (m2.xlarge) computing nodes (http://aws.amazon.com/ec2/). We implemented existing mining framework [35,39,40], UF-growth [32], tube-growth [33], BLIMP-growth [34], and our data analytic algorithm MrCloud all in the Java programming language. The stock version of Apache Hadoop 0.20.0 was used.…”

Section: Evaluation Resultsmentioning

confidence: 99%

“…With this notion of expected support, existing tree-based algorithms-such as UF-growth [32], tube-growth [33] and BLIMP-growth [34]-mine frequent patterns from uncertain data by first scanning the uncertain database once to compute the expected support of all domain items (i.e., singleton patterns). Infrequent items are pruned as their extensions/supersets are guaranteed to be infrequent.…”

Section: Uncertain Data Miningmentioning

confidence: 99%

“…Infrequent items are pruned as their extensions/supersets are guaranteed to be infrequent. The algorithms then scan the database a second time to insert all transactions (with only frequent items) into a tree (e.g., UF-tree [32], TPC-tree [33], or BLIMP-tree [34]). Each node in the tree captures (i) an item x; (ii) its existential probability P (x, t j ); and (iii) its occurrence count.…”

Section: Uncertain Data Miningmentioning

confidence: 99%

“…With this notion, each item in a transaction t j in databases containing precise data can be viewed as an item with a 100% likelihood of being present in t j . To find frequent patterns from uncertain data, several uncertain data mining algorithms (e.g., UF-growth [32], tube-growth [33], and BLIMP-growth [34] algorithms) have been proposed. For many real-life applications, users look for all frequent patterns.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Data Analytic Algorithm for Managing, Querying, and Processing Uncertain Big Data in Cloud Environments

Jiang

Leung

2015

Algorithms

View full text Add to dashboard Cite

Big data are everywhere as high volumes of varieties of valuable precise and uncertain data can be easily collected or generated at high velocity in various real-life applications. Embedded in these big data are rich sets of useful information and knowledge. To mine these big data and to discover useful information and knowledge, we present a data analytic algorithm in this article. Our algorithm manages, queries, and processes uncertain big data in cloud environments. More specifically, it manages transactions of uncertain big data, allows users to query these big data by specifying constraints expressing their interests, and processes the user-specified constraints to discover useful information and knowledge from the uncertain big data. As each item in every transaction in these uncertain big data is associated with an existential probability value expressing the likelihood of that item to be present in a particular transaction, computation could be intensive. Our algorithm uses the MapReduce model on a cloud environment for effective data analytics on these uncertain big data. Experimental results show the effectiveness of our data analytic algorithm for managing, querying, and processing uncertain big data in cloud environments.

show abstract

Section: Evaluation Resultsmentioning

confidence: 86%