Rank: 78137 based on 0 downloads (last 30 days) and 0 file submitted
photo

musi

E-mail

Personal Profile:

I am a PhD student

Professional Interests:
recommender systems, Information retrieval

 

Watch this Author's files

 

Comments and Ratings by musi View all
  • 1 - 3 of
  • 3
Updated File Comments Rating
24 Aug 2011 k-means++ Cluster multivariate data using the k-means++ algorithm. Author: Laurent S

Hi thanks for it.
I am just confused how we will choose the next centroid.
Let's suppose we have chosen c1 (first centroid) uniformly at random. I calculate prob of each point with c1 and it is {p1,p2,....pn}.

What is the criteria to choose the next point? e.g. shld I choose the the one with max prob? min prob? or just with random prob (see next)

for all point i \in n
if(rand (0.0,1.0) < p(i))
c2= p(i)

Thanks
16 Jul 2011 barweb (BARgraph With Error Bars) Plots a fully customizable grouped bar graph with error bars Author: Bolu Ajiboye

(The above statement is to put samples values associated with each bar in the x-axis)
16 Jul 2011 barweb (BARgraph With Error Bars) Plots a fully customizable grouped bar graph with error bars Author: Bolu Ajiboye

Pretty great:

it is working

% groupnames
{'1000', '2000', '5000', '10000', '15000', '20000', '25730'},

Contact us