Detecting Changes in Rare Patterns from Data Streams

Huang, David Tse Jung; Koh, Yun Sing; Dobbie, Gillian; Pears, Russel

doi:10.1007/978-3-319-06605-9_36

Cited by 14 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Up to now, a number of RPM algorithms have been extensively proposed, such as CFP-growth++ [17]. Among them, several studies are designed to deal with dynamic data streams [10,13,14]. There are numerous candidates for most algorithms based on Apriori mechanism.…”

Section: Rare Pattern Miningmentioning

confidence: 99%

FRI-Miner: Fuzzy Rare Itemset Mining

Cui

Gan

Lin

et al. 2021

Preprint

View full text Add to dashboard Cite

Data mining is a widely used technology for various real-life applications of data analytics and is important to discover valuable association rules in transaction databases. Interesting itemset mining plays an important role in many real-life applications, such as market, e-commerce, finance, and medical treatment. To date, various data mining algorithms based on frequent patterns have been widely studied, but there are a few algorithms that focus on mining infrequent or rare patterns. In some cases, infrequent or rare itemsets and rare association rules also play an important role in real-life applications. In this paper, we introduce a novel fuzzy-based rare itemset mining algorithm called FRI-Miner, which discovers valuable and interesting fuzzy rare itemsets in a quantitative database by applying fuzzy theory with linguistic meaning. Additionally, FRI-Miner utilizes the fuzzy-list structure to store important information and applies several pruning strategies to reduce the search space. The experimental results show that the proposed FRI-Miner algorithm can discover fewer and more interesting itemsets by considering the quantitative value in reality. Moreover, it significantly outperforms state-of-the-art algorithms in terms of effectiveness (w.r.t. different types of derived patterns) and efficiency (w.r.t. running time and memory usage).

show abstract

Section: Rare Pattern Miningmentioning

confidence: 99%

FRI-Miner: Fuzzy Rare Itemset Mining

Cui

Gan

Lin

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…2.2.2 Temporal shape approaches. e temporal shape is another unique property of time-series that can be exploited in segmentation [10,18] where changes in the temporal shape pa erns of a time-series were used to estimate the segment boundaries. FLOSS, Fast Low-Cost Semantic Segmentation [10] works under the principle that pa erns of similar shape were each associated with the same segment class and occur within close temporal proximity of each other.…”

Section: Time-series Segmentation Approachesmentioning

confidence: 99%

“…e limitations of such assumptions were described in the Introduction section. In contrast to FLOSS, which is based on the most similar repeated pa erns, the authors in [18] proposed a segmentation model based on rare temporal pa erns. Although shape-based methods can be bene cial for time-series composed of repeated shape pa erns, performance will degrade when segments are composed of diverse shapes or when the shape pa erns of a segment dri over time.…”

Section: Time-series Segmentation Approachesmentioning

confidence: 99%

ESPRESSO: Entropy and ShaPe awaRe timE-Series SegmentatiOn for processing heterogeneous sensor data

Deldari,

Smith,

Sadri

et al. 2020

Preprint

View full text Add to dashboard Cite

Extracting informative and meaningful temporal segments from high-dimensional wearable sensor data, smart devices, or IoT data is a vital preprocessing step in applications such as Human Activity Recognition (HAR), trajectory prediction, gesture recognition, and lifelogging. In this paper, we propose ESPRESSO (Entropy and ShaPe awaRe timE-Series SegmentatiOn), a hybrid segmentation model for multi-dimensional time-series that is formulated to exploit the entropy and temporal shape properties of time-series. ESPRESSO di ers from existing methods that focus upon particular statistical or temporal properties of time-series exclusively. As part of model development, a novel temporal representation of time-series W CAC was introduced along with a greedy search approach that estimate segments based upon the entropy metric. ESPRESSO was shown to o er superior performance to four state-of-the-art methods across seven public datasets of wearable and wear-free sensing. In addition, we undertake a deeper investigation of these datasets to understand how ESPRESSO and its constituent methods perform with respect to di erent dataset characteristics. Finally, we provide two interesting case-studies to show how applying ESPRESSO can assist in inferring daily activity routines and the emotional state of humans.

show abstract

“…Nevertheless, there is much room for expansion in this area when it comes to mining of rare patterns from data streams. Both single pass [58,59,128] and multiple pass algorithms [53,83] have been designed for generation of rare patterns from data stream. The techniques developed, however, cannot efficiently handle the issue of concept drift.…”

Section: Mining Sparse Data and Datasets With Long Patternsmentioning

confidence: 99%

Rare pattern mining: challenges and future perspectives

Borah

Nath

2018

Complex Intell. Syst.

View full text Add to dashboard Cite

Extracting frequent patterns from databases has always been an imperative task for the data mining community. Literature has endowed plentiful endeavors to this research area with significant breakthroughs. Mining of rare patterns although being subsided has proved to be of vital importance in many domains. The application of rare pattern mining is inevitable and thus has become an emerging field of research. However, discovering rare patterns from databases comes up with numerous challenges. This article provides a concise overview of the various research issues involved in rare pattern mining through experimental analysis using real-life and synthetic datasets. Rare pattern mining being a new area has abundant scope and there are still certain gaps that need to be filled. In this article, we also present some viable future directions for the researchers to better perceive the area of rare pattern mining. KeywordsRare patterns • Frequent patterns • Rare item dilemma • Challenges B Anindita Borah

show abstract

Detecting Changes in Rare Patterns from Data Streams

Cited by 14 publications

References 13 publications

FRI-Miner: Fuzzy Rare Itemset Mining

FRI-Miner: Fuzzy Rare Itemset Mining

ESPRESSO: Entropy and ShaPe awaRe timE-Series SegmentatiOn for processing heterogeneous sensor data

Rare pattern mining: challenges and future perspectives

Contact Info

Product

Resources

About