SavedAdd your journal

Journals / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi / 2020 / Cilt: 35 - Sayı: 1

Kd-tree and adaptive radius (KD-AR Stream) based real-time data stream clustering

K-boyutlu ağaç ve uyarlanabilir yarıçap (KD-AR Stream) tabanlı gerçek zamanlı akan veri kümeleme

Ali ŞENOL
Hacer KARACAN
Pages
337–354
Publication Date
2019-10-25
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results showed that KD-AR Stream algorithm has a good clustering performance within a reasonable time by comparison with the other algorithms.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Data stream clustering is one of the most popular topics of today's world where the amount of data reaches incredible levels in parallel with technological developments.
📊
Key Findings & Data:The most important problems encountered in data stream clustering approaches are the fact that most of the approaches consists of an online and offline phases, the definition of the number of cluster, or the need to set a limitation to this number, the problems encountered in determining optimum radius value, and the problems encountered in concept evolution.
💡
Conclusion & Impact:The results showed that KD-AR Stream algorithm has a good clustering performance within a reasonable time by comparison with the other algorithms.
🔬Design:Research Article
📍Scope:Comprehensive Sample
⚙️Variables:Akan veri kümeleme · k-boyutlu ağaç · uyarlanabilir yarıçap

Abstract

Data stream clustering is one of the most popular topics of today's world where the amount of data reaches incredible levels in parallel with technological developments. The most important problems encountered in data stream clustering approaches are the fact that most of the approaches consists of an online and offline phases, the definition of the number of cluster, or the need to set a limitation to this number, the problems encountered in determining optimum radius value, and the problems encountered in concept evolution. The present study proposes an evolutionary based solution method, which is based on Kd-Tree and adaptive radius (KD-AR Stream) to perform real-time clustering on the streaming data. The proposed approach has been compared with SE-Stream, DPStream and CEDAS algorithms in terms of both cluster quality and execution time. The results showed that KD-AR Stream algorithm has a good clustering performance within a reasonable time by comparison with the other algorithms.

Özet

Akan veri kümeleme, teknolojik gelişmelere paralel olarak veri miktarının inanılmaz boyutlara ulaştığı gümünüzün popüler konularından biridir. Akan veri kümeleme yaklaşımlarında karşılaşılan en önemli problemler çoğu yaklaşımın çevrimiçi ve çevrimdışı evreden oluşması, küme sayısını tanımlama veya bu sayıya bir sınır koyma zorunluluğu, en doğru yarıçap değerini belirlemede yaşanan problemler ve önerilen modellerin kendisini gelen yeni verilere adapte etmesinde (concept evolution) yaşanan problemlerdir. Bu problemlerin yanında, neredeyse bu alandaki bütün çalışmaların sayısal miktar tabanlı bir özetleme yapması da bazı uygulamalar için ihtiyacı karşılamamaktadır. Oysa son 1 saniyede veya son 1 saatte gelen veriler şeklinde çalışan zaman tabanlı bir özetleme yaklaşımına da ihtiyaç vardır. Bu çalışmada, K-boyutlu ağaç, uyarlanabilir yarıçap tabanlı (KD-AR Stream) ve kümeleme adaptasyonu özelliğine sahip gerçek zamanlı akan verileri kümeleyen bir yaklaşım önerilmektedir. Önerdiğimiz yöntem SE-Stream, DPStream ve CEDAS algoritmaları ile hem kümeleme başarısı hem de işlem performansı açısından karşılaştırılmıştır. Elde edilen sonuçlar KD-AR Stream algoritmasının diğer algoritmalara göre yüksek bir kümeleme başarısını makul bir sürede gerçekleştirdiğini göstermektedir.

🔖

Cite This Article

ŞENOL, A., KARACAN, H. (2019). Kd-tree and adaptive radius (KD-AR Stream) based real-time data stream clustering. Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, Cilt: 35 - Sayı: 1, 337–354. https://ajindex.com/articles/19849

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (11)

  1. Antonellis, P., C. Makris, and N. Tsirakis, Algorithms for clustering clickstream data. Information Processing Letters, 2009. 109(8): p. 381-385.
  2. Yin, C., L. Xia, and J. Wang. Application of an Improved Data Stream Clustering Algorithm in Intrusion Detection System. in Advanced Multimedia and Ubiquitous Engineering. 2017. Singapore: Springer Singapore. Yin, C., L. Xia, and J. Wang. Data Stream Clustering Algorithm Based on Bucket Density for Intrusion Detection. in Advances in Computer Science and Ubiquitous Computing. 2018. Singapore: Springer Singapore. Li, Z.Q., A New Data Stream Clustering Approach about Intrusion Detection. Advanced Materials Research, 2014. 926-930: p. 2898-2901.
  3. Hendricks, D., Using real-time cluster configurations of streaming asynchronous features as online state descriptors in financial markets. Pattern Recognition Letters, 2017. 97: p. 21-28.
  4. Aggarwal, C.C., Data Streams: An Overview and Scientific Applications, in Scientific Data Mining and Knowledge Discovery: Principles and Foundations, M.M. Gaber, Editor. 2010, Springer Berlin Heidelberg: Berlin, Heidelberg. p. 377-397.
  5. King, R.C., et al., Application of data fusion techniques and technologies for wearable health monitoring. Medical Engineering & Physics, 2017. 42: p. 1-12.
  6. Gravina, R., et al., Multi-sensor fusion in body sensor networks: State-of-the-art and research challenges. Information Fusion, 2017. 35: p. 68-80.
  7. Manzi, A., P. Dario, and F. Cavallo, A Human Activity Recognition System Based on Dynamic Clustering of Skeleton Data. Sensors (Basel, Switzerland), 2017. 17(5): p. 1100. Diaz-Rozo, J., C. Bielza, and P. Larrañaga, Clustering of Data Streams with Dynamic Gaussian Mixture Models. An IoT Application in Industrial Processes. IEEE Internet of Things Journal, 2018: p. 1-1.
  8. Tasnim, S., et al. Semantic-Aware Clustering-based Approach of Trajectory Data Stream Mining. in 2018 International Conference on Computing, Networking and Communications (ICNC). 2018.
  9. Ankleshwaria, T.B. and J.S. Dhobi, Mining Data Streams: A Survey. International Journal of Advance Research in Computer Science and Management Studies, 2014. 2(2): p. 379-386.
  10. Ikonomovska, E., S. Loskovska, and D. Gjorgjevik, A survey of stream data mining, in Eighth International Conference with International Participation – ETAI 2007. 2007:
  11. Ohrid, Republic of Macedonia. Şenol, A. and Karacan H., A Survey on Data Stream Clustering Techniques. European Journal of Science and Technology, 2018(13): p. 17-30. Aggarwal