diff --git a/demo/kaggle-higgs/README.md b/demo/kaggle-higgs/README.md
index a3c208002..b3db23266 100644
--- a/demo/kaggle-higgs/README.md
+++ b/demo/kaggle-higgs/README.md
@@ -16,5 +16,6 @@ make
 
 
 
-
+Speed
+=====
 speedtest.py compares xgboost's speed on this dataset with sklearn.GBM
diff --git a/demo/multiclass_classification/README.md b/demo/multiclass_classification/README.md
new file mode 100644
index 000000000..4d66ee06a
--- /dev/null
+++ b/demo/multiclass_classification/README.md
@@ -0,0 +1,8 @@
+Demonstrating how to use XGBoost accomplish Multi-Class classification task on [UCI Dermatology dataset](https://archive.ics.uci.edu/ml/datasets/Dermatology)
+
+1. Run runexp.sh
+```bash
+./runexp.sh
+```
+
+Explainations can be found in [wiki](https://github.com/tqchen/xgboost/wiki)
diff --git a/demo/multiclass_classification/runexp.sh b/demo/multiclass_classification/runexp.sh
new file mode 100755
index 000000000..0af814725
--- /dev/null
+++ b/demo/multiclass_classification/runexp.sh
@@ -0,0 +1,9 @@
+#!/bin/bash
+if [ -f dermatology.data ]
+then
+    echo "use existing data to run multi class classification"
+else
+    echo "getting data from uci, make sure you are connected to internet"
+    wget https://archive.ics.uci.edu/ml/machine-learning-databases/dermatology/dermatology.data
+fi
+python train.py
diff --git a/demo/multiclass_classification/train.py b/demo/multiclass_classification/train.py
new file mode 100755
index 000000000..38d818890
--- /dev/null
+++ b/demo/multiclass_classification/train.py
@@ -0,0 +1,42 @@
+#! /usr/bin/python
+import sys
+import numpy as np
+sys.path.append('../../python/')
+import xgboost as xgb
+
+# label need to be 0 to num_class -1
+data = np.loadtxt('./dermatology.data', delimiter=',',converters={33: lambda x:int(x == '?'), 34: lambda x:int(x)-1 } )
+sz = data.shape
+
+train = data[:int(sz[0] * 0.7), :]
+test = data[int(sz[0] * 0.7):, :]
+
+train_X = train[:,0:33]
+train_Y = train[:, 34]
+
+
+test_X = test[:,0:33]
+test_Y = test[:, 34]
+
+xg_train = xgb.DMatrix( train_X, label=train_Y)
+xg_test = xgb.DMatrix(test_X, label=test_Y)
+# setup parameters for xgboost
+param = {}
+# use softmax multi-class classification
+param['objective'] = 'multi:softmax'
+# scale weight of positive examples
+param['bst:eta'] = 0.1
+param['bst:max_depth'] = 6
+param['silent'] = 1
+param['nthread'] = 4
+param['num_class'] = 6
+
+watchlist = [ (xg_train,'train'), (xg_test, 'test') ]
+num_round = 5
+bst = xgb.train(param, xg_train, num_round, watchlist );
+# get prediction
+pred = bst.predict( xg_test );
+
+print 'predicting, classification error=%f' % (sum( int(pred[i]) != test_Y[i] for i in xrange(len(test_Y))) / float(len(test_Y)) )
+
+
diff --git a/regrank/xgboost_regrank.h b/regrank/xgboost_regrank.h
index b2649735c..e3868002a 100644
--- a/regrank/xgboost_regrank.h
+++ b/regrank/xgboost_regrank.h
@@ -97,8 +97,8 @@ namespace xgboost{
             */
             inline void InitTrainer(void){
                 if( mparam.num_class != 0 ){
-                    if( name_obj_ != "softmax" ){
-                        name_obj_ = "softmax";
+                    if( name_obj_ != "multi:softmax" ){
+                        name_obj_ = "multi:softmax";
                         printf("auto select objective=softmax to support multi-class classification\n" );
                     }
                 }