87
edits
Seminar on Internet Technologies (Winter 2017/2018): Difference between revisions
Seminar on Internet Technologies (Winter 2017/2018) (view source)
Revision as of 15:15, 21 September 2017
, 21 September 2017→Topics
(→Topics) |
(→Topics) |
||
| Line 53: | Line 53: | ||
Clustering is the unsupervised learning algorithm which groups unlabeled data into similar sub-groups. The clustering problem has been addressed in many contexts (social network, structure biological network ..). In this topic, we review and compare different approach address this problem. There are two main “small topics”: | Clustering is the unsupervised learning algorithm which groups unlabeled data into similar sub-groups. The clustering problem has been addressed in many contexts (social network, structure biological network ..). In this topic, we review and compare different approach address this problem. There are two main “small topics”: | ||
a, Non-model based algorithms: Kmeans, spectral clustering, DBSCAN .. | a, Non-model based algorithms: Kmeans, spectral clustering, DBSCAN .. | ||
b, A probabilistic model-based algorithm: Expectation Maximization, Gibbs sampler for Gaussian mixture model | b, A probabilistic model-based algorithm: Expectation Maximization, Gibbs sampler for Gaussian mixture model. | ||
There are some useful practical parts which help students apply algorithms in real data. | There are some useful practical parts which help students apply algorithms in real data. | ||
| Thach Nguyen (Chuong-Thach.Nguyen@mpibpc.mpg.de) | | Thach Nguyen (Chuong-Thach.Nguyen@mpibpc.mpg.de) | ||