best mining algorithm to cluster the follwoing dataset in matlab

1 visualizzazione (ultimi 30 giorni)
i have excel table contain data-set , the follwoing example explain portion of my table
in my data set i have 100 folder and these folder contain some content suppose i consider only image files , i need to cluster this folders based on those images and the number of cluster each time differ when execute the code , as follow
folder1 folder2 folder3 folder4 folder5 .......... folder100
car 1 1 0 0 0 0
orange 0 0 1 1 0 0
apple 0 0 1 1 0 0
vehicle 1 1 0 0 0 0
flight 1 0 0 1 1 0
airplan 1 0 0 1 0 0
bicycle 1 1 0 0 0 0
motor 1 1 0 0 0 0
grapes 0 0 1 0 0 0
...
the tested images are 20 for example and usr can add or changed nd every time this table content are changed but for this result hope to get clusters as follow :
cluster1( motors) include ( folder 1 , folder2 , ...etc )
cluster2( fruite) include ( folder3 , folder4 , ... etc )
and so on ,... etc
what are the best mining algorithm to find clusters at each execution time , consider number of cluster each time differ.

Risposta accettata

Walter Roberson
Walter Roberson il 9 Set 2015
The best data mining algorithm for that purpose has not been invented yet and probably involves mathematical concepts that no-one has dreamed of yet. What we do know, though, is that it will require less than 100,000 billion years to run.
If you want practical data mining algorithms then you should be researching them in textbooks and science papers such as through http://scholar.google.com . MATLAB Answers is about MATLAB, not about image content recognition.

Più risposte (0)

Categorie

Scopri di più su Statistics and Machine Learning Toolbox in Help Center e File Exchange

Prodotti

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by