Skip to content

Ulrich777/Kaggle-Challenge-Kernels-Methods-For-Machine-Learning

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Kaggle-Challenge-Kernels-Methods-For-Machine-Learning

The purpose of this challenge was to implement and gain practical experience with the kernel methods seen during the class. We were challenged to predict whether a DNA sequence is a binding site to a specific transcription factor. Beside reaching the highest possible score, we explored classical kernel methods along with the kernels for biological sequences described in class We also tried to take into account some research topics encountered in class, namely how to combine several kernels In our approach. We simultaneously learned the optimal parameters of SVM and the optimal polynomial combination of our available kernels . We used three kinds of kernels: spectrum kernels, mismatch kernels and exponentially smoothed spectrum kernels.

We were ranked 8/76

About

No description or website provided.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages