- 1. Clustering Clustering A survey about some of the clustering techniques
- 2. BIG DATA BIG DATA •Everyday a huge amount of data is produced ,Too much data to be processed by man .So the help of computers is needed
- 3. Clustering Clustering •Cluster analysis or clustering is the task of grouping a set of objects in such a way that objects in the same group are more similar to each other than to those in other groups
- 4. Clustering Applications Clustering Applications •Aviation/Astronomy: Astronomical data •Biology : Multiple gene expression-Identification Of functionally related genes •Climate : Discovery of climate indices •Energy : Discovering energy consumption pattern •Finance : Finding seasonality patterns •Medicine : Detecting brain activity •Etc
- 5. Clustering Algorithms Clustering Algorithms while There is a lot of clustering Algorithms not all of them can be used in all situations ,depending on the purpose of the clustering type of data ,and many more factors. certain types of algorithms may give better results than others. Choosing the right algorithm is very essential to achieve the desired results
- 6. K-means K-means •K-means algorithm is one of the most used algorithm In clustering ,where a certain number of cluster centers are chosen randomly, then the algorithm works on the minimization of the Total distance between all Objects in a cluster from their cluster center.
- 7. Relocation Algorithm Relocation Algorithm •In the relocation algorithm we start with an initial clustering having a certain number of clusters X , then For each data point compute the dissimilarity matrix and store all resultant matrices computed for all points , Finally find a new cluster that will be better in terms of a certain criterion ,the new cluster is found by swapping two members between two different clusters
- 8. Agglomerative hierarchical clustering Agglomerative hierarchical clustering A hierarchical clustering method works by grouping data objects into a tree of clusters. Two types of hierarchical clustering methods are often distinguished: agglomerative and divisive depending upon whether a bottom-up or top-down strategy is followed
- 9. Agglomerative Hierarchical Clustering Agglomerative Hierarchical Clustering The agglomerative hierarchical clustering method is more popular than the divisive method. It starts by placing each object in its own cluster and then merges these atomic clusters into larger and larger clusters, until all the objects are in a single cluster or until certain termination conditions are satisfied
- 10. Divisive Hierarchical Clustering Divisive Hierarchical Clustering Similar to Agglomerative Hierarchical Clustering except that the algorithm starts by collecting all the data points into a single cluster then start dividing it into smaller and smaller clusters until we have a certain amount of clusters
- 11. Self-organizing maps(SOM) Self-organizing maps(SOM) Self-organizing maps are a class of neural networks structure and trained by an iterative unsupervised or self-organizing procedure . Each training-iteration consists of three steps: the presentation of a randomly chosen input vector from the input space, the evaluation of the network, and an update of the weight vectors
- 12. Summary Summary There are many clustering algorithms for different situations and for different types of data ,and each will achieve a different purpose but the sure thing is that clustering is very powerful tool that can be utilized for a plethora of applications

Thank you for your comment.