[Spark-10994] Add local clustering coefficient computation in GraphX

[SherlockYang]

The local clustering coefficient of a vertex (node) in a graph quantifies how close its neighbours are to being a clique (complete graph).

More specifically, the local clustering coefficient C_i for a vertex v_i is given by the proportion of links between the vertices within its neighbourhood divided by the number of links that could possibly exist between them.

Duncan J. Watts and Steven Strogatz introduced the measure in 1998 to determine whether a graph is a small-world network.

Usage

Here is a usage example for LocalClusteringCoefficient:

import org.apache.spark.graphx._
import org.apache.spark._

val conf = new SparkConf().setAppName("testApp")
val sc = new SparkContext(conf)
// load a graph
val graph = GraphLoader.edgeListFile(sc, "graph.txt").partitionBy(PartitionStrategy.RandomVertexCut)

// perform the local clustering coefficient computation 
val LccCounter = graph.localClusteringCoefficient()

// output results for each vertex
val verts = LccCounter.vertices
verts.collect().foreach { case (vid, count) =>
    println(vid + ": " + count)
}