K-Means Clustering based on MapReduce

K-Means is a clustring algorithm designed to partition unlabelled data into a certain number (that’s the “K”) of distinct groupings. K-Means clustering based on MapReduce speeds up the execution either on standalone or distributed computing. MapReduce has several advantages for speeding up K-means clustring algorithm:

• Parallel: it makes the best of multicore computer.
• Simpilicity: only consider Map and Reduce concepts.
• Distribution: Makes this algorithm possible to run on distributed mode.

Reference

• K-means: A Complete Introduction, online available: https://towardsdatascience.com/
• A Very Brief Introduction of MapReduce, online available: https://hci.standford.edu/courses/cs448g
• davide-cocoomini/KMeans-MapReduce, online available: https://github.com/davide-cocoomini/KMeans-MapReduce
• IRIS Flower Dataset, online available: https://www.kaggle.com/arshid/iris-flower-dataset
• Weather Dateset, online available: https://www.kaggle.com/prakharrathi25/weather-data-clustering-using-k-means/notebook

Contribution

• Hao-Ying Cheng (Masker Tim)
  • Email: t109598001@ntut.org.tw
  • Affliation: National Taipei University of Technology
• Yueh Tang
  • Email: t109598033@ntut.org.tw
  • Affliation: National Taipei University of Technology