Testing= the=C2=A0columnSimilarity method in Spark, I create a RowMatrix = object:

```val temp =3D sc.parallelize(Array((5.=
0, 1.0, 4.0), (2.0, 3.0, 8.0),  (4.0, 5.0, 10.0<=
/span>), (1.0,=
3.0, 6.0)))val row=
s =3D temp.map(line =3D> {  Vectors.dense(Array(line._1, line._2, line._3))})val mat =3D new RowMatrix(rows)```

the matrix is:
5=C2=A0 1=C2=A0 =C2=A04
2=C2=A0 3=C2=A0 =C2=A08
4=C2=A0 5=C2=A0 =C2=A010
1=C2= =A0 3=C2=A0 =C2=A06

It will return the cosinSimila= rity of rows:
(5, 2, 4, 1)
(1, 3, 5, 3)
(4, 8= , 10, 6)
that is :

MatrixEntry(0,2,= 0.8226366627527562)
MatrixEntry(0,1,0.755742181606458)
= MatrixEntry(1,2,0.9847319278346619)

The prob= lem is when I set threshold:
`val est =3D mat.columnSimilarities(0.5)`
and the result of some pairs will be = greater than one and because it's similarity the result should be betwe= en zero and one!

MatrixEntry(0,2,2.8217416025= 43195)
MatrixEntry(0,1,1.319846878608914)

My primary question is what is the interpretation of results greate= r than one?
Does Spark use the=C2=A0DIMSUM= =C2=A0algorithm for just cosinSimilarities with a threshold or it use=C2=A0= DIMSUM=C2=A0for similarities without a threshold, too?
<= /div>
--00000000000079eef6057ab7c661--