Predictions for Kaggle Bike sharing prediction using R

r
kaggle

#1

Given data set i have applied both Random Forest and SVM.But Random Forest is giving more accuracy than SVM. I want the reason behind it?


#2

@gangadhar_chintha123

It would help, if you can provide some details on features you have used / created.

In general, it is hard to predict which algorithm would do better over other with out understanding the data and features, if both the algorithms are available.

Regards,
Kunal


#3

hello sir,

In my data set totally i have 12 features and 10886 observations. About the features datetime,season, holiday,workingday,weather temp,atemp,humidity,windspeed ,casual,registered,count(causal+registered)


In the test set we have to predict the count.Firstly i have done preprocess on training data and then apply the Random Forest algoritham and SVM.But Random Forest is giving more accuracy.if u need any information i can give.


#4

SVM is usually applied for a classification problem. This is a regression problem.

Kunal


#5

SVM can also be applicable for regression.In data set some features are having categorical and some features are having continuous values. Thats why i have applied SVM.


#6

have you tried rattle package in R. you can try a lot of algorithms in a single click. It is a GUI. see http://rattle.togaware.com