Spectrum Data Science and Machine Learning Internship Project Report

-> Data Science and Machine Learning Internship at Spectrum Club CET,Bhubaneswar.This is a work from home basis Internship provided by Spectrum Club, CET College.

-> In this Internship program we have done data preprocessing, data visualization,data cleaning, data manipulation and prediction by using Machine Learning Algorithm.

-> Different Libraries used for the project is:

1. Numpy

2. Pandas

3. Matplotlib

4. seaborn

5. sklearn

6. Statsmodel.api

-> Task- 1 : In this task a dictionary was given and we have to perform operation on this dictionary by using Numpy library and perform data visualization by using Matplotlib library and perform some statistical operations( Mean,Median,Max,Min).

-> Task- 2 : In this task a "Student" dataset was given and we perform several operations-

            1. Creating Dataframe from CSV file
           
            2. Creating a new column from existing columns and deleting columns
           
            3. Handling categorical values
           
            4. Data Visualization : Scatter Plot, Box Plot

-> Task- 3 : In this task we have use Machine Learning along with several operations-

            1. Conversion of nominal values into numeric values using LabelEncoder/OneHotEncoder
            
            2. Convert the data into Training and Testing data
            
            3. Model Selection
            
            4. Calculate the accuracy of model
            
            5. use backward elimination method to find the most important attributes 
            
            6. Use of different Machine Learning Models (Decision Tree, Random Forest)

1. Linear Regression

   Results:

      Train Score:  0.97542
   
      Test Score:  0.9628
   
      Predict Score:  0.9976
   

2. Decision Tree Regression

   Results:

      Train Score:  1.0
   
     Test Score:  0.9504
   
     Predict Score:  1.0
   

3. Random Forest Regression

   Results:

     Train Score: 0.9961
   
    Test Score:  0.9693
   
    Predict Score:  0.9973
   

Future Scope : We can deploy the model so that it take inputs and predict the output.

I have created a web interface using Flask and deployed the model in Heroku Cloud.

Check it out-https://studentprediction-api.herokuapp.com/

Name - Nitish Kumar Nayak

Regd No. - 1805106016

Branch - MCA

CET , Bhubaneswar