Содержание
- 2. Compare results K-means: http://www.naftaliharris.com/blog/visualizing-k-means-clustering/ (I’ll choose -> Gaussian Mixture, Smiley Face) DBScan: http://www.naftaliharris.com/blog/visualizing-dbscan-clustering/ (Gaussian Mixture, Smiley
- 3. DBSCAN Density-based spatial clustering of applications with noise (DBSCAN) is a data clustering algorithm proposed by
- 4. Preliminary Consider a set of points in some space to be clustered. For the purpose of
- 5. Preliminary If p is a core point, then it forms a cluster together with all points
- 6. Preliminary Two points p and q are density-connected if there is a point o such that
- 7. Example In this diagram, minPts = 3. Point A and the other red points are core
- 8. Algorithm DBSCAN requires two parameters: ε (eps) and the minimum number of points required to form
- 9. Algorithm If a point is found to be a dense part of a cluster, its ε-neighborhood
- 10. Main procedure
- 11. Procedure expandCluster
- 12. Procedure regionQuery
- 13. Note The algorithm can be simplified by merging the per-point "has been visited" and "belongs to
- 14. Complexity DBSCAN visits each point of the database, possibly multiple times (e.g., as candidates to different
- 15. Parameter estimation (minPts) Ideally, minPts is the desired minimum cluster size. Otherwise a minimum minPts can
- 16. Parameter estimation (ε) If ε is chosen much too small, a large part of the data
- 17. Advantages DBSCAN does not require to specify the number of clusters in the data a priori,
- 18. Disadvantages DBSCAN is not entirely deterministic: border points that are reachable from more than one cluster
- 19. DBScan and DBMS Nothing…
- 21. Скачать презентацию