477
click to vote
PKDD
2015
Springer
10 years 2 months ago
2015
Springer
Most hypothesis testing in machine learning is done using the frequentist null-hypothesis significance test, which has severe drawbacks. We review recent Bayesian tests which over...
168
click to vote
PKDD
2015
Springer
10 years 2 months ago
2015
Springer
Abstract. We propose a probabilistic method for inferring the geographical locations of linked objects, such as users in a social network. Unlike existing methods, our model does n...
125
Voted
PKDD
2015
Springer
10 years 2 months ago
2015
Springer
We present a solution to the MoReBikeS challenge in ECML PKDD 2015 conference by analysing data from different aspects, by visualising latent patterns, by building a set of featur...
134
click to vote
PKDD
2015
Springer
10 years 2 months ago
2015
Springer
The goal of domain adaptation is to solve the problem of di↵erent joint distribution of observation and labels in the training and testing data sets. This problem happens in many...