Merge pull request #165 from ayushi424/main

Campus Placement Analysis & Prediction

Author: prathimacode-hub

Machine Learning/Campus Placement Analysis & Prediction/Dataset/Placement_Data_Full_Class.csv

@@ -0,0 +1,49 @@
+# **Campus Placement Analysis & Prediction**
+
+**GOAL**
+
+To analyze various factors and predict salary offered to candidates during campus placements using machine learning algorithm.
+
+**DATASET**
+
+Dataset can be downloaded from [here](https://www.kaggle.com/benroshan/factors-affecting-campus-placement).
+
+**WHAT I HAD DONE**
+- Step 1: Data Preprocessing & Exploration
+- Step 2: Data Training & Model Creation
+- Step 3: Performance Evaluation
+
+
+**Screenshots**
+
+![](https://github.com/ayushi424/PyAlgo-Tree/blob/main/Machine%20Learning/Campus%20Placement%20Analysis%20%26%20Prediction/Images/cp1.jpg)
+![](https://github.com/ayushi424/PyAlgo-Tree/blob/main/Machine%20Learning/Campus%20Placement%20Analysis%20%26%20Prediction/Images/cp2.jpg)
+![](https://github.com/ayushi424/PyAlgo-Tree/blob/main/Machine%20Learning/Campus%20Placement%20Analysis%20%26%20Prediction/Images/cp3.jpg)
+![](https://github.com/ayushi424/PyAlgo-Tree/blob/main/Machine%20Learning/Campus%20Placement%20Analysis%20%26%20Prediction/Images/cp4.jpg)
+**MODEL USED**
+-  Decision Tree Regressor
+
+**LIBRARIES NEEDED**
+- pandas
+- numpy
+- matplotlib
+- seaborn
+- sklearn (For data traning, importing models and performance check)
+
+
+**Accuracy of different models used**
+ - By using Decision Tree Regressor model 
+ ```python
+    Accuracy achieved :  1.00
+ ``` 
+
+
+
+**CONCLUSION**
+
+Performance of Decision tree regressor is highyly efficient.
+
+
+**Author** 
+
+[Ayushi Shrivastava](https://github.com/ayushi424)

Machine Learning/Campus Placement Analysis & Prediction/Images/cp1.jpg

@@ -0,0 +1,127 @@
+# -*- coding: utf-8 -*-
+"""campus_placement_analysis&prediction.ipynb
+
+Automatically generated by Colaboratory.
+
+Original file is located at
+    https://colab.research.google.com/drive/1H8LaqWZcR6OqEhPmavKEbh2duuiTLypC
+
+# **Campus Placement Analysis & Prediction**
+
+Here, in this project we will analyze various features that affect campus placements and then perform prediction using decision tree regressor algorithm.
+
+Following steps are followed:
+
+-  **Data preprocessing and exploration** to understand what kind of data will we working on.
+-  **Data Training** using train-test-split method from sklearn to split the data into training and testing data & Model Creation using decision tree regressor algorithm.
+-  **Performance Evaluation** by error and accuracy check to find how efficient algorithm is for this project.
+
+For the dataset being used in this project [click here](https://www.kaggle.com/benroshan/factors-affecting-campus-placement)
+
+### **Data Preprocessing & Exploration**
+"""
+
+#importing pandas library.
+import pandas as pd
+
+#loading and reading data through following
+data=pd.read_csv('/content/Placement_Data_Full_Class.csv')
+data
+
+#to view shape of the dataset i.e. total number of rows and columns.
+data.shape
+
+#to view first 5 rows of the dataset.
+data.head()
+
+#to view last 5 rows of the dataset.
+data.tail()
+
+#to view different columns of the dataset.
+data.columns
+
+#to view memory usage, non-null values, datatypes of columns.
+data.info()
+
+#to view statistical summary of the dataset.
+data.describe()
+
+#to check for any missing or null values in the dataset.
+data.isnull().sum()
+
+"""There are 67 null values in 'salary', we can't proceed with this.
+
+We need to replace the null values by the mean of that respective column.
+"""
+
+data['salary'].isnull().sum()
+
+data['salary'].mean()
+
+data['salary'].fillna('288655',inplace=True)
+# fillna fucntion will fill the null values(where null=TRUE) with the mean value.
+
+data['salary'].isnull().sum()
+
+#to check again for any missing or null values in the dataset.
+data.isnull().sum()
+
+#to view total null values in the dataset.
+data.isnull().sum().sum()
+
+"""Now that the dataset has no null values i.e. dataset is cleaned and proper.
+
+We can now proceed with further steps.
+
+### **Data Training**
+"""
+
+data.info()
+
+data1=data.drop(['gender','ssc_b','hsc_b','hsc_s','degree_t','workex','specialisation','status'],axis=1)
+data1.info()
+
+#converting data into int datatype to avoid errors below.
+prepareddata=data1.astype(int)
+prepareddata.head()
+
+# Import train_test_split from sklearn.model_selection
+from sklearn.model_selection import train_test_split
+# Here, X is the data which will have feature and y will have our target.
+x=prepareddata.drop(['salary'],axis=1)  
+y=prepareddata['salary']
+
+# Split data into training data and testing data
+x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2,random_state=500) 
+#Ratio used for splitting training and testing data is 8:2 respectively
+
+"""### **Model Creation using Decision Tree Regressor Algorithm**"""
+
+# Importing decision tree regressor 
+from sklearn.tree import DecisionTreeRegressor 
+reg = DecisionTreeRegressor()
+
+#Fitting data into the model.
+reg.fit(x_train, y_train)
+
+# Making predictions on Test data 
+pred = reg.predict(x_test)
+
+pred
+
+"""### **Performance Evaluation**"""
+
+import numpy as np
+from sklearn.metrics import mean_squared_error
+print("Model\t\t\t          RootMeanSquareError  \t\t    Accuracy of the model") 
+print("""Decision Tree Regressor \t\t {:.2f} \t \t\t {:.2f}""".format(  np.sqrt(mean_squared_error(y_test, pred)), reg.score(x_train,y_train)))
+
+"""Conclusion Drawn:
+*  Accuracy of the decision tree regressor model for this project is 1.00 which is an excellent accuracy.
+
+*  Decision tree regressor is a highly efficient model and widely used for regression tasks,various prediction projects etc.
+
+**Author**
+
+[Ayushi Shrivastava](https://github.com/ayushi424)
+"""