BDP 05: CLUSTERING OF LARGE UNLABELED DATASETS

Source Code

1D Clustering

• K-Means Clustering on the Ages of the Users dataset.

2D Clustering

• K-Means Clustering on two features of the Posts dataset (Score and ViewCount)

2D Clustering with Join of Datasets

• Includes the helper MapReduce used to preprocess the data and join the datasets.
• K-Means Clustering of the User Ages and Badges Count of the Users and Badges dataset.

Data Preprocessing

• Helper MapReduce used to normalise tha data points.

Extract Sample From Posts

• Helper MapReduce used to extract sample data from the Posts dataset for plot purposes.