Implicit Heterogeneous Features Embedding in Deep Knowledge Tracing决策树实现

Implicit Heterogeneous Features Embedding in Deep Knowledge Tracing决策树实现

论文
Implicit Heterogeneous Features Embedding in Deep Knowledge Tracing
Haiqin Yang1 ·Lap Pong Cheung2

数据集下载地址(需要科学的上网)
link

我已经下好了的数据集和论文的下载链接，点这里

代码

import pandas as pd
import numpy as np
from sklearn.tree import DecisionTreeClassifier
from sklearn.tree import export_graphviz

# 读取指定列索引字段的数据
# usedclos=['original', 'attempt_count','ms_first_response','answer_text','assistment_id','type','hint_count','hint_total',
#           'overlap_time','first_action','opportunity','tutor_mode','correct']

usedclos=['original', 'attempt_count','ms_first_response','assistment_id','hint_count','hint_total',
          'overlap_time','first_action','opportunity','correct']
csv_data = pd.read_csv("./skill_builder_data.csv", usecols=usedclos,encoding='utf-8')
print(csv_data.columns)
Y=csv_data['correct']
#删除correct列
del csv_data['correct']
print(csv_data.columns)

X=csv_data

feature =csv_data.columns
classname =['incorrect','correct']
#max_depth是最大层数
tree_clf = DecisionTreeClassifier(max_depth=4)
tree_clf.fit(X, Y)

export_graphviz(
            tree_clf,
            out_file=("DecisionTreeClassifier.dot"),
            feature_names=feature,
            class_names=classname,
            rounded=True,
            filled=True,

        )
print('成功！')
'''
apt-get install graphviz 
生成决策树的命令
dot -Tpng DecisionTreeClassifier.dot -o 123.png
'''

实现结果