Github user srowen commented on a diff in the pull request:
https://github.com/apache/spark/pull/15795#discussion_r86744662
--- Diff: docs/ml-features.md ---
@@ -1396,3 +1396,134 @@ for more details on the API.
{% include_example python/ml/chisq_selector_example.py %}
</div>
</div>
+
+# Locality Sensitive Hashing
+[Locality Sensitive
Hashing(LSH)](https://en.wikipedia.org/wiki/Locality-sensitive_hashing) is a
class of dimension reduction hash families, which can be used as both feature
transformation and machine-learned ranking. Difference distance metric has its
own LSH family class in `spark.ml`, which can transform feature columns to hash
values as new columns. Besides feature transforming, `spark.ml` also
implemented approximate nearest neighbor algorithm and approximate similarity
join algorithm using LSH.
+
+In this section, we call a pair of input features a false positive if the
two features are hashed into the same hash bucket but they are far away in
distance, and we define false negative as the pair of features when their
distance are close but they are not in the same hash bucket.
+
+## Random Projection for Euclidean Distance
+**Note:** Please note that this is different than the [Random Projection
for cosine
distance](https://en.wikipedia.org/wiki/Locality-sensitive_hashing#Random_projection).
+
+[Random
Projection](https://en.wikipedia.org/wiki/Locality-sensitive_hashing#Stable_distributions)
is the LSH family in `spark.ml` for Euclidean distance. The Euclidean distance
is defined as follows:
+`\[
+d(\mathbf{x}, \mathbf{y}) = \sqrt{\sum_i (x_i - y_i)^2}
+\]`
+Its LSH family projects features onto a random unit vector and divide the
projected results to hash buckets:
+`\[
+h(\mathbf{x}) = \lfloor \frac{\mathbf{x} \cdot \mathbf{v}}{r} \rfloor
+\]`
+where `v` is a normalized random unit vector and `r` is user-defined
bucket length.
+
+The input features in Euclidean space are represented in vectors. Both
sparse and dense vectors are supported.
+
+The bucket length can be used to trade off the performance of random
projection. A larger bucket lowers the false negative rate but usually
increases running time and false positive rate.
+
+<div class="codetabs">
+<div data-lang="scala" markdown="1">
+
+Refer to the [RandomProjection Scala
docs](api/scala/index.html#org.apache.spark.ml.feature.RandomProjection)
+for more details on the API.
+
+{% include_example
scala/org/apache/spark/examples/ml/RandomProjectionExample.scala %}
+</div>
+
+<div data-lang="java" markdown="1">
+
+Refer to the [RandomProjection Java
docs](api/java/org/apache/spark/ml/feature/RandomProjection.html)
+for more details on the API.
+
+{% include_example
java/org/apache/spark/examples/ml/JavaRandomProjectionExample.java %}
+</div>
+</div>
+
+## MinHash for Jaccard Distance
+[MinHash](https://en.wikipedia.org/wiki/MinHash) is the LSH family in
`spark.ml` for Jaccard distance where input features are sets of natural
numbers. Jaccard distance of two sets is defined by the cardinality of their
intersection and union:
+`\[
+d(\mathbf{A}, \mathbf{B}) = 1 - \frac{|\mathbf{A} \cap
\mathbf{B}|}{|\mathbf{A} \cup \mathbf{B}|}
+\]`
+As its LSH family, MinHash applies a random [perfect hash
function](https://en.wikipedia.org/wiki/Perfect_hash_function) `g` to each
elements in the set and take the minimum of all hashed values:
+`\[
+h(\mathbf{A}) = \min_{a \in \mathbf{A}}(g(a))
+\]`
+
+Input sets for MinHash is represented in vectors which dimension equals
the total number of elements in the space. Each dimension of the vectors
represents the status of an elements: zero value means the elements is not in
the set; non-zero value means the set contains the corresponding elements. For
example, `Vectors.sparse(10, Array[(2, 1.0), (3, 1.0), (5, 1.0)])` means there
are 10 elements in the space. This set contains elem 2, elem 3 and elem 5.
--- End diff --
Same sort of comment: isn't the intuition that MinHash approximates the
Jaccard similarity without actually computing it completely?
This may again reduce to my different understanding of the technique, but
MinHash isn't a hashing technique per se. it relies on a given family of hash
functions to approximate set similarity. I'm finding it a little hard to view
it as an 'LSH family'
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
