[SPARK-10595] [ML] [MLLIB] [DOCS] Various ML guide cleanups

[jkbradley]

Various ML guide cleanups.

• ml-guide.md: Make it easier to access the algorithm-specific guides.
• LDA user guide: EM often begins with useless topics, but running longer generally improves them dramatically. E.g., 10 iterations on a Wikipedia dataset produces useless topics, but 50 iterations produces very meaningful topics.
• mllib-feature-extraction.html#elementwiseproduct: “w” parameter should be “scalingVec”
• Clean up Binarizer user guide a little.
• Document in Pipeline that users should not put an instance into the Pipeline in more than 1 place.
• spark.ml Word2Vec user guide: clean up grammar/writing
• Chi Sq Feature Selector docs: Improve text in doc.

CC: @mengxr @feynmanliang

[SparkQA]

Test build #42436 has finished for PR 8752 at commit 53d757a.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • [ChiSqSelector](api/scala/index.html#org.apache.spark.mllib.feature.ChiSqSelector) implements Chi-Squared feature selection. It operates on labeled data with categorical features.ChiSqSelectororders features based on a Chi-Squared test of independence from the class, and then filters (selects) the top features which the class label depends on the most. This is akin to yielding the features with the most predictive power.

[feynmanliang]

Not sure if we should include model summaries in this description; I had a mailing list question about where that feature is documented

[jkbradley]

Yeah, there's not a great place. I'll try sticking a note here.

[jkbradley]

@feynmanliang Thanks for reviewing. Just updated per your comments.

[feynmanliang]

The classname is backticked in ChiSqSelector but not here or in Binarizer, we should choose one and be consistent. I would vote for backticking everything since that's what I've been doing

[feynmanliang]

LGTM after changes

[SparkQA]

Test build #42500 has finished for PR 8752 at commit 91f4edd.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • and then filters (selects) the top features which the class label depends on the most.

[mengxr]

Merged into master. Thanks!