density based method and expectation maximization

density based method and expectation maximization

• 1.
  k.Sivapriya Msc(IT) Nadar saraswathi collegeof arts and science
• 2.
  Density based methods Clusters – dense regions of objects Low density regions – Noise DBSCAN Density Based Spatial Clustering of Applications with  Noise OPTICS Ordering Points To Identify the Clustering Structure DENCLUE DEN sity Based CLU stEring
• 3.
  DBSCAN Cluster – maximalset of density connected points  Grows regions with sufficiently high density into clusters  e-neighborhood  MinPts and Core object  Directly Density Reachable An object p is directly density reachable from object q if p is within the e-neighborhood of q and q is a core Object
• 4.
  DBSCAN  Density Reachable An object p is density reachable from q, if there is  a chain of objects p1, …pn, p1=q and pn=p such that  pi+1 is directly density reachable from pi
• 5.
  Continue…..  Density Connected: Anobject p is density connected to object q if there is an object o such that both p and q are density reachable from o.
• 6.
  Conti….. Arbitrarily select apoint p  Retrieve all points density-reachable from p  If p is a core point, a cluster is formed.  If p is a border point, no points are density-reachable from p, then DBSCAN visits the next point of the database.  Continue the process until all of the points have been processed.  Complexity : O(n log n) / O(n2)
• 7.
  OPTICS: A Cluster-OrderingMethod OPTICS: Ordering Points To Identify the Clustering Structure  Produces a special order of the database with respect to its density-based clustering structure  Good for both automatic and interactive cluster analysis, including finding intrinsic clustering structure  Can be represented graphically or using visualization techniques
• 8.
  OPTICS In DBSCAN, fora constant MinPts value, density based clusters with respect to a higher density (lower value of e) are completely contained in lower density sets.  DBSCAN is extended so that Objects are processed in a specific order.  Selects an object that is density-reachable with respect to loweste value  Core distance of an object p : smallest e’ value that makes {p} a core object  Reachability distance of an object q with respect to p = max (core-distance of p, d(p,q))8