algorithm - How to build up the observation array in Accord.NET using C# -


i'm trying learn ropes in accord.net , ai world... goal clustering list of customers using k-means algorithm. each customer, got 3 features:

customerid, productcategory, totqty, totamount  aaa, 01, 50, 3000 aaa, 02, 10, 150 bbb, 01, 45, 2700 ... 

now, have pass observations k-means algorithm:

double[][] observations = ... (?)  // create new k-means algorithm kmeans kmeans = new kmeans(k: 10);  // compute , retrieve data centroids var clusters = kmeans.learn(observations);  // use centroids parition data int[] labels = clusters.decide(observations); 

first question: have group data customer? this:

double[][] observation = {                    new double[] { 1, 50, 3000,  2, 10, 150 },   new double[] { 1, 45, 2700} } 

or:

double[][] observation = {                    new double[] { 1, 50, 3000},   new double[] { 2, 10, 150},   new double[] { 1, 45, 2700} } 

second question: how trace result original customerid? mean got result assigned label int[] labels = clusters.decide(observations); how can determine customer belongs cluster/label?

i made generic k-means library c#

so can use second question. (after got centroid can objects belong centroid)

https://github.com/pashkovdenis/k-means/


Comments

Popular posts from this blog

Is there a better way to structure post methods in Class Based Views -

performance - Why is XCHG reg, reg a 3 micro-op instruction on modern Intel architectures? -

jquery - Responsive Navbar with Sub Navbar -