cluster analysis - data mining project Dilemma -
i research set of data, consisting of 2 data files: first contains user id id artists , ranking of users artists want rank. second data file contains id , name artists
i have chosen research question is: artist popular or not?
in other words,by given new singer, not found in data file, using algorithms, classify artist , know if popular or not.
for prediction step chose use logistic regression method problem earlier. not know how, technically, determine existing data defined successful artist unsuccessful.
i thought of methods, example:k-means k=2 (but in method have problem function disance),knn k=2 etc.
i need guidance ,refers how make clustering existing data , general tips project. thank you.
Comments
Post a Comment