uniform Effect of K-means Clustering
up vote
0
down vote
favorite
In the following link is discussed the uniform Effect of K-means Clustering:
https://www.springer.com/cda/content/document/cda_downloaddocument/9783642298066-c2.pdf?SGWID=0-0-45-1338325-p174318763
He used the following equation:
$ frac{displaystyle2 d(C_1,C_2)}{displaystyle n_1n_2}=frac{displaystyle d(C_1,C_1)}{displaystyle n_1^2}+frac{displaystyle d(C_2,C_2)}{displaystyle n_2^2}+2||m_1−m_2||^2.$
where
$ d(C_1,C_2)=sum_{x_iin C_1}sum_{x_jin C_2}||x_i−x_j||^2$ with $ |C_1|=n_1$ and $|C_2 |=n_2$.
I tried to prove this, but I do not get this. He uses the fact that
$ d(C_1,C_1)=2(n_1-1)sum_{i=1}^{n_1} ||x_i||^2 -4 sum_{1leq i < j leq n_1}langle x_i,x_jrangle.$
$ d(C_2,C_2)=2(n_2-1)sum_{i=1}^{n_2} ||y_i||^2 -4 sum_{1leq i < j leq n_2}langle y_i,y_jrangle.$
$ d(C_1,C_2)=2n_2sum_{i=1}^{n_1} ||x_i||^2+2n_1sum_{i=1}^{n_2} ||y_i||^2 -4 sum_{1leq ileq n_1}sum_{1leq jleq n_2}langle x_i,y_jrangle.$
And furthermore with
$||m_1−m_2||^2=langle sum_{i=1}^{n_1}x_i/n_1-sum_{j=1}^{n_1}y_j/n_2,sum_{i=1}^{n_1}x_i/n_1-sum_{j=1}^{n_1}y_j/n_2 rangle = frac{1}{displaystyle n_1^2}sum_{i=1}^{n_1} ||x_i||^2 + frac{2}{displaystyle n_1^2}sum_{1leq i < jleq n_1} langle x_i, x_jrangle +frac{1}{displaystyle n_2^2}sum_{i=1}^{n_2} ||y_i||^2 + frac{2}{displaystyle n_2^2}sum_{1leq i < jleq n_2} langle y_i, y_jrangle - frac{2}{n_1n_2} sum_{1leq ileq n_1}sum_{1leq jleq n_2}langle x_i,y_jrangle, $
I obtain that
$ frac{displaystyle d(C_1,C_2)}{displaystyle n_1n_2}=frac{displaystyle d(C_1,C_1)}{displaystyle n_1^2}+frac{displaystyle d(C_2,C_2)}{displaystyle n_2^2}+2||m_1−m_2||^2.$
Where is my error?
analysis vector-analysis inner-product-space machine-learning clustering
add a comment |
up vote
0
down vote
favorite
In the following link is discussed the uniform Effect of K-means Clustering:
https://www.springer.com/cda/content/document/cda_downloaddocument/9783642298066-c2.pdf?SGWID=0-0-45-1338325-p174318763
He used the following equation:
$ frac{displaystyle2 d(C_1,C_2)}{displaystyle n_1n_2}=frac{displaystyle d(C_1,C_1)}{displaystyle n_1^2}+frac{displaystyle d(C_2,C_2)}{displaystyle n_2^2}+2||m_1−m_2||^2.$
where
$ d(C_1,C_2)=sum_{x_iin C_1}sum_{x_jin C_2}||x_i−x_j||^2$ with $ |C_1|=n_1$ and $|C_2 |=n_2$.
I tried to prove this, but I do not get this. He uses the fact that
$ d(C_1,C_1)=2(n_1-1)sum_{i=1}^{n_1} ||x_i||^2 -4 sum_{1leq i < j leq n_1}langle x_i,x_jrangle.$
$ d(C_2,C_2)=2(n_2-1)sum_{i=1}^{n_2} ||y_i||^2 -4 sum_{1leq i < j leq n_2}langle y_i,y_jrangle.$
$ d(C_1,C_2)=2n_2sum_{i=1}^{n_1} ||x_i||^2+2n_1sum_{i=1}^{n_2} ||y_i||^2 -4 sum_{1leq ileq n_1}sum_{1leq jleq n_2}langle x_i,y_jrangle.$
And furthermore with
$||m_1−m_2||^2=langle sum_{i=1}^{n_1}x_i/n_1-sum_{j=1}^{n_1}y_j/n_2,sum_{i=1}^{n_1}x_i/n_1-sum_{j=1}^{n_1}y_j/n_2 rangle = frac{1}{displaystyle n_1^2}sum_{i=1}^{n_1} ||x_i||^2 + frac{2}{displaystyle n_1^2}sum_{1leq i < jleq n_1} langle x_i, x_jrangle +frac{1}{displaystyle n_2^2}sum_{i=1}^{n_2} ||y_i||^2 + frac{2}{displaystyle n_2^2}sum_{1leq i < jleq n_2} langle y_i, y_jrangle - frac{2}{n_1n_2} sum_{1leq ileq n_1}sum_{1leq jleq n_2}langle x_i,y_jrangle, $
I obtain that
$ frac{displaystyle d(C_1,C_2)}{displaystyle n_1n_2}=frac{displaystyle d(C_1,C_1)}{displaystyle n_1^2}+frac{displaystyle d(C_2,C_2)}{displaystyle n_2^2}+2||m_1−m_2||^2.$
Where is my error?
analysis vector-analysis inner-product-space machine-learning clustering
add a comment |
up vote
0
down vote
favorite
up vote
0
down vote
favorite
In the following link is discussed the uniform Effect of K-means Clustering:
https://www.springer.com/cda/content/document/cda_downloaddocument/9783642298066-c2.pdf?SGWID=0-0-45-1338325-p174318763
He used the following equation:
$ frac{displaystyle2 d(C_1,C_2)}{displaystyle n_1n_2}=frac{displaystyle d(C_1,C_1)}{displaystyle n_1^2}+frac{displaystyle d(C_2,C_2)}{displaystyle n_2^2}+2||m_1−m_2||^2.$
where
$ d(C_1,C_2)=sum_{x_iin C_1}sum_{x_jin C_2}||x_i−x_j||^2$ with $ |C_1|=n_1$ and $|C_2 |=n_2$.
I tried to prove this, but I do not get this. He uses the fact that
$ d(C_1,C_1)=2(n_1-1)sum_{i=1}^{n_1} ||x_i||^2 -4 sum_{1leq i < j leq n_1}langle x_i,x_jrangle.$
$ d(C_2,C_2)=2(n_2-1)sum_{i=1}^{n_2} ||y_i||^2 -4 sum_{1leq i < j leq n_2}langle y_i,y_jrangle.$
$ d(C_1,C_2)=2n_2sum_{i=1}^{n_1} ||x_i||^2+2n_1sum_{i=1}^{n_2} ||y_i||^2 -4 sum_{1leq ileq n_1}sum_{1leq jleq n_2}langle x_i,y_jrangle.$
And furthermore with
$||m_1−m_2||^2=langle sum_{i=1}^{n_1}x_i/n_1-sum_{j=1}^{n_1}y_j/n_2,sum_{i=1}^{n_1}x_i/n_1-sum_{j=1}^{n_1}y_j/n_2 rangle = frac{1}{displaystyle n_1^2}sum_{i=1}^{n_1} ||x_i||^2 + frac{2}{displaystyle n_1^2}sum_{1leq i < jleq n_1} langle x_i, x_jrangle +frac{1}{displaystyle n_2^2}sum_{i=1}^{n_2} ||y_i||^2 + frac{2}{displaystyle n_2^2}sum_{1leq i < jleq n_2} langle y_i, y_jrangle - frac{2}{n_1n_2} sum_{1leq ileq n_1}sum_{1leq jleq n_2}langle x_i,y_jrangle, $
I obtain that
$ frac{displaystyle d(C_1,C_2)}{displaystyle n_1n_2}=frac{displaystyle d(C_1,C_1)}{displaystyle n_1^2}+frac{displaystyle d(C_2,C_2)}{displaystyle n_2^2}+2||m_1−m_2||^2.$
Where is my error?
analysis vector-analysis inner-product-space machine-learning clustering
In the following link is discussed the uniform Effect of K-means Clustering:
https://www.springer.com/cda/content/document/cda_downloaddocument/9783642298066-c2.pdf?SGWID=0-0-45-1338325-p174318763
He used the following equation:
$ frac{displaystyle2 d(C_1,C_2)}{displaystyle n_1n_2}=frac{displaystyle d(C_1,C_1)}{displaystyle n_1^2}+frac{displaystyle d(C_2,C_2)}{displaystyle n_2^2}+2||m_1−m_2||^2.$
where
$ d(C_1,C_2)=sum_{x_iin C_1}sum_{x_jin C_2}||x_i−x_j||^2$ with $ |C_1|=n_1$ and $|C_2 |=n_2$.
I tried to prove this, but I do not get this. He uses the fact that
$ d(C_1,C_1)=2(n_1-1)sum_{i=1}^{n_1} ||x_i||^2 -4 sum_{1leq i < j leq n_1}langle x_i,x_jrangle.$
$ d(C_2,C_2)=2(n_2-1)sum_{i=1}^{n_2} ||y_i||^2 -4 sum_{1leq i < j leq n_2}langle y_i,y_jrangle.$
$ d(C_1,C_2)=2n_2sum_{i=1}^{n_1} ||x_i||^2+2n_1sum_{i=1}^{n_2} ||y_i||^2 -4 sum_{1leq ileq n_1}sum_{1leq jleq n_2}langle x_i,y_jrangle.$
And furthermore with
$||m_1−m_2||^2=langle sum_{i=1}^{n_1}x_i/n_1-sum_{j=1}^{n_1}y_j/n_2,sum_{i=1}^{n_1}x_i/n_1-sum_{j=1}^{n_1}y_j/n_2 rangle = frac{1}{displaystyle n_1^2}sum_{i=1}^{n_1} ||x_i||^2 + frac{2}{displaystyle n_1^2}sum_{1leq i < jleq n_1} langle x_i, x_jrangle +frac{1}{displaystyle n_2^2}sum_{i=1}^{n_2} ||y_i||^2 + frac{2}{displaystyle n_2^2}sum_{1leq i < jleq n_2} langle y_i, y_jrangle - frac{2}{n_1n_2} sum_{1leq ileq n_1}sum_{1leq jleq n_2}langle x_i,y_jrangle, $
I obtain that
$ frac{displaystyle d(C_1,C_2)}{displaystyle n_1n_2}=frac{displaystyle d(C_1,C_1)}{displaystyle n_1^2}+frac{displaystyle d(C_2,C_2)}{displaystyle n_2^2}+2||m_1−m_2||^2.$
Where is my error?
analysis vector-analysis inner-product-space machine-learning clustering
analysis vector-analysis inner-product-space machine-learning clustering
edited 7 hours ago
asked 7 hours ago
Patricio
16711
16711
add a comment |
add a comment |
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3006724%2funiform-effect-of-k-means-clustering%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown