Scribd Logo
Upload

Welcome to Scribd!

  • Palash Bhai - Machine Learning Assignment

    Uploaded by

    PalashKulshrestha
    269 views18 pages
    The document provides details on analyzing and modeling a political dataset. It includes: 1. Descriptive statistics and checks for null values which found no missing data but some duplicate rows. 2. Univariate and bivariate analysis along with outlier detection was performed. 3. Several models were applied including logistic regression, LDA, KNN, naive bayes, and random forest. Random forest performed best with 99.7% accuracy on training data and 89.6% on test data. 4. The models were compared and random forest was determined to be the best/most optimized model based on AUC scores.

    Original Description:

    ML Assignment

    Original Title

    Palash Bhai_Machine Learning Assignment

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    The document provides details on analyzing and modeling a political dataset. It includes: 1. Descriptive statistics and checks for null values which found no missing data but some duplicate rows. 2. Univariate and bivariate analysis along with outlier detection was performed. 3. Several models were applied including logistic regression, LDA, KNN, naive bayes, and random forest. Random forest performed best with 99.7% accuracy on training data and 89.6% on test data. 4. The models were compared and random forest was determined to be the best/most optimized model based on AUC scores.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    269 views18 pages

    Palash Bhai - Machine Learning Assignment

    Uploaded by

    PalashKulshrestha
    The document provides details on analyzing and modeling a political dataset. It includes: 1. Descriptive statistics and checks for null values which found no missing data but some duplicate rows. 2. Univariate and bivariate analysis along with outlier detection was performed. 3. Several models were applied including logistic regression, LDA, KNN, naive bayes, and random forest. Random forest performed best with 99.7% accuracy on training data and 89.6% on test data. 4. The models were compared and random forest was determined to be the best/most optimized model based on AUC scores.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    of 18

    Problem 1

    1. Read the dataset. Do the descriptive statistics and do null value condition check. Write an inference
    on it.

    Head of the data is as follows

    Information about the data type of various columns is given below

    Description of the data set is given below. Few interpretations are as follows:

    1. Mean and 50% data representation for Blair, Hague, Europe and Political knowledge has
    significant difference
    2. Count for all columns is same, therefore, there are no missing values
    Checking for null values to confirm our interpretation

    There are 8 duplicate rows in the dataset, however, they cannot be inferred as the age for various
    duplicate items is different so we will not eliminate these rows

    2. Perform Univariate and Bivariate Analysis. Do exploratory data analysis. Check for Outliers.
    Checking for outlier
    Data Preparation

    1. Encode the data (having string values) for Modelling. Is Scaling necessary here or not? Data Split: Split
    the data into train and test (70:30).

    Modelling

    1. Apply Logistic Regression and LDA (linear discriminant analysis).


    Logistic Regression

    Model accuracy score for training data– 83.97%

    AUC for training data – 88.9%

    Model score for training data – 82.31%

    AUC for training data – 88.9%

    Confusion Matrix for training data


    Confusion matrix for test data
    Linear Discriminant Analysis
    AUC for the Training Data: 88.9%

    AUC for the Test Data: 88.4%


    2. Apply KNN Model and Naïve Bayes Model. Interpret the results.

    KNN Model

    AUC for the Training Data: 91.1%

    AUC for the Test Data: 86.1%


    KNN_train_precision: 0.87

    KNN_train_recall: 0.88

    KNN_train_f1: 0.88

    KNN_test_precision: 0.86

    KNN_test_recall: 0.87

    KNN_test_f1: 0.86

    Naïve Bayes Model

    AUC for the Training Data: 88.6%

    AUC for the Test Data: 88.5%


    3. Model Tuning, Bagging (Random Forest should be applied for Bagging) and Boosting.
    AUC for the Training Data: 99.7%

    AUC for the Test Data: 89.6%


    4. Performance Metrics: Check the performance of Predictions on Train and Test sets using
    Accuracy, Confusion Matrix, Plot ROC curve and get ROC_AUC score for each model. Final
    Model: Compare the models and write inference which model is best/optimized.

    Based on the various results of the models we can infer that Random Forest has the most optimized
    model as it has the highest data as per Area under Curve.

    Particulars Training Data Test Data


    Logistical Regression 83.9% 82.37%
    LDA 88.9% 88.4%
    KNN 91.1% 86.1%
    Naives Bayes 88.6% 88.5%
    Random Forest 99.7% 89.6%

    Inference

    1. Based on these predictions, what are the insights?

    Accuracy on Test data is 99% and on Train data is 80%.

    AUC is greater than 89% for both.

    Recall and Precision is low and same on both data.

    While the model results between training and test sets are similar, indicating no under or overfitting
    issues.

    Problem 2
    1. Find the number of characters, words and sentences for the mentioned documents.

    For 1941 Roosevelt speech

    The number of sentences in the text are: 68


    The number of words in the text are: 1,360

    The number of characters in the text are: 7,571

    For 1961 Kennedy speech

    The number of sentences in the text are: 52

    The number of words in the text are: 1,390

    The number of characters in the text are: 7,618

    For 1973 Nixon speech

    The number of sentences in the text are: 68

    The number of words in the text are: 1,819

    The number of characters in the text are: 9,991

    2. Remove all the stopwords from all the three speeches.

    3. Which word occurs the most number of times in his inaugural address for each president?
    Mention the top three words. (after removing the stopwords)
    Top three words for Roosevelt are: nation, know, spirit

    Top three words for Kennedy are: let, us, world

    Top three words for Nixon are: us, let, America

    4. Plot the word cloud of each of the speeches of the variable. (after removing the stopwords)

    Word cloud for Roosevelt

    Word cloud for Kennedy after cleaning


    Word cloud for Nixon after cleaning

    You might also like