July 2017
Beginner to intermediate
715 pages
17h 3m
English
Manual evaluation means looking at the results manually and using the domain expertise to assess the quality of clusters and if they make any sense.
The manual check is usually done in the following way:
When looking at the data, we want to ask ourselves the following questions:
If the answer to both the question is yes, then the clustering results are good. Additionally, the way we sample data is also important. For example, in case of K-means, we should sample some items that are close to the centroid, and some items ...
Read now
Unlock full access