MOA is an open source framework for Big Data stream mining. It includes a collection of machine learning algorithms (classification, regression, clustering, outlier detection, concept drift detection and recommender systems) and tools for evaluation.
Hey! I was diving into the source code of
moa/moa/src/main/java/moa/clusterers/clustream/WithKmeans.java, and found out that it would be really useful if right after the user calls getClusteringResult(), the macro clusters returned have a unique id (current id is -1 for all of them), and then when the user calls getMicroClusteringResult(), these ones have a coherent id too, representing the former labeling (current micro clusters id is -1 for all of them) ...
It seems that the moa/moa/src/main/java/moa/cluster/Cluster.java setId() method is not used and I do not know if that is on purpose (like for some reason).
Having the micro clusters labeled would be useful for calculating metrics to see how well the clustering resulted, for plotting different groups with different colors, and so on.
If you think what I remarked is correct, I will submit a PR.
Hey! I was diving into the source code of
moa/moa/src/main/java/moa/clusterers/clustream/WithKmeans.java
, and found out that it would be really useful if right after the user callsgetClusteringResult()
, the macro clusters returned have a unique id (current id is -1 for all of them), and then when the user callsgetMicroClusteringResult()
, these ones have a coherent id too, representing the former labeling (current micro clusters id is -1 for all of them) ...It seems that the
moa/moa/src/main/java/moa/cluster/Cluster.java
setId() method is not used and I do not know if that is on purpose (like for some reason).Having the micro clusters labeled would be useful for calculating metrics to see how well the clustering resulted, for plotting different groups with different colors, and so on.
If you think what I remarked is correct, I will submit a PR.
Any feedback is welcomed. Thanks!!