diff options
| -rw-r--r-- | docs/ml-features.md | 14 | ||||
| -rw-r--r-- | docs/mllib-feature-extraction.md | 14 |
2 files changed, 20 insertions, 8 deletions
diff --git a/docs/ml-features.md b/docs/ml-features.md index a39b31c8f7..a7f710fa52 100644 --- a/docs/ml-features.md +++ b/docs/ml-features.md @@ -1331,10 +1331,16 @@ for more details on the API. ## ChiSqSelector `ChiSqSelector` stands for Chi-Squared feature selection. It operates on labeled data with -categorical features. ChiSqSelector orders features based on a -[Chi-Squared test of independence](https://en.wikipedia.org/wiki/Chi-squared_test) -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. +categorical features. ChiSqSelector uses the +[Chi-Squared test of independence](https://en.wikipedia.org/wiki/Chi-squared_test) to decide which +features to choose. It supports three selection methods: `KBest`, `Percentile` and `FPR`: + +* `KBest` chooses the `k` top features according to a chi-squared test. This is akin to yielding the features with the most predictive power. +* `Percentile` is similar to `KBest` but chooses a fraction of all features instead of a fixed number. +* `FPR` chooses all features whose false positive rate meets some threshold. + +By default, the selection method is `KBest`, the default number of top features is 50. User can use +`setNumTopFeatures`, `setPercentile` and `setAlpha` to set different selection methods. **Examples** diff --git a/docs/mllib-feature-extraction.md b/docs/mllib-feature-extraction.md index 353d391249..87e1e027e9 100644 --- a/docs/mllib-feature-extraction.md +++ b/docs/mllib-feature-extraction.md @@ -225,10 +225,16 @@ features for use in model construction. It reduces the size of the feature space both speed and statistical learning behavior. [`ChiSqSelector`](api/scala/index.html#org.apache.spark.mllib.feature.ChiSqSelector) implements -Chi-Squared feature selection. It operates on labeled data with categorical features. -`ChiSqSelector` orders 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. +Chi-Squared feature selection. It operates on labeled data with categorical features. ChiSqSelector uses the +[Chi-Squared test of independence](https://en.wikipedia.org/wiki/Chi-squared_test) to decide which +features to choose. It supports three selection methods: `KBest`, `Percentile` and `FPR`: + +* `KBest` chooses the `k` top features according to a chi-squared test. This is akin to yielding the features with the most predictive power. +* `Percentile` is similar to `KBest` but chooses a fraction of all features instead of a fixed number. +* `FPR` chooses all features whose false positive rate meets some threshold. + +By default, the selection method is `KBest`, the default number of top features is 50. User can use +`setNumTopFeatures`, `setPercentile` and `setAlpha` to set different selection methods. The number of features to select can be tuned using a held-out validation set. |