[
https://issues.apache.org/jira/browse/FLINK-1735?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14532547#comment-14532547
]
Till Rohrmann commented on FLINK-1735:
--------------------------------------
I guess that a tokenizer/sentence splitter makes much sense if we want to do
text classification. If you want to, then you can open a JIRA issue and
implement it.
I guess that a Seq[T] would be most generic as an input for the feature hasher,
right. If that is not possible, then you can start with Set[String] and see how
to generalize it later. I'm currently reworking the pipelining and this will
allow you to define specialized implementations for different input types (e.g.
String, Image, ...)
> Add FeatureHasher to machine learning library
> ---------------------------------------------
>
> Key: FLINK-1735
> URL: https://issues.apache.org/jira/browse/FLINK-1735
> Project: Flink
> Issue Type: New Feature
> Components: Machine Learning Library
> Reporter: Till Rohrmann
> Assignee: Felix Neutatz
> Labels: ML
>
> Using the hashing trick [1,2] is a common way to vectorize arbitrary feature
> values. The hash of the feature value is used to calculate its index for a
> vector entry. In order to mitigate possible collisions, a second hashing
> function is used to calculate the sign for the update value which is added to
> the vector entry. This way, it is likely that collision will simply cancel
> out.
> A feature hasher would also be helpful for NLP problems where it could be
> used to vectorize bag of words or ngrams feature vectors.
> Resources:
> [1] [https://en.wikipedia.org/wiki/Feature_hashing]
> [2]
> [http://scikit-learn.org/stable/modules/feature_extraction.html#feature-extraction]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
