exam-sub 1.0.0 → 1.0.2

package/data/mlf.txt

@@ -1,26 +1,32 @@
- Q1 to Q6: Short, Simple and Easy-to-Understand Solutions
-Question 1: Sales Data EDA
-Aim: Perform Exploratory Data Analysis on sales data to understand patterns, missing values and unusual transactions.
-1. Import libraries and load data
+================================================================================
+Q1: SALES DATA EXPLORATORY DATA ANALYSIS (EDA)
+================================================================================
+
+AIM:
+Perform Exploratory Data Analysis on sales data to understand patterns, missing 
+values, and unusual transactions.
+
+PYTHON PROGRAM:
+--------------------------------------------------------------------------------
 import pandas as pd
 import numpy as np
 import matplotlib.pyplot as plt
 import seaborn as sns
+
+# 1. Load Data and Inspect
 df = pd.read_csv("sales_data.csv")
 print(df.head())
 print(df.shape)
 print(df.info())
 
-2. Convert Date
+# 2. Convert Date
 df['Date'] = pd.to_datetime(df['Date'])
 print(df['Date'].dtype)
 
-3. Identify variable types
+# 3. Identify Variable Types
 print(df.dtypes)
-Typical types: Order_ID – integer/object; Date – DateTime; Product, Category, Region, Customer_Type – categorical;
-Quantity, Unit_Price, Discount, Sales – numerical.
 
-4. Check and 5. Handle missing values
+# 4 & 5. Check and Handle Missing Values
 print(df.isnull().sum())
 df['Quantity'] = df['Quantity'].fillna(df['Quantity'].median())
 df['Unit_Price'] = df['Unit_Price'].fillna(df['Unit_Price'].median())
@@ -29,102 +35,110 @@ df['Sales'] = df['Sales'].fillna(df['Sales'].median())
 df['Region'] = df['Region'].fillna(df['Region'].mode()[0])
 df['Category'] = df['Category'].fillna(df['Category'].mode()[0])
 
-6. Univariate analysis
+# 6. Univariate Analysis
 sns.histplot(df['Sales'], kde=True)
 plt.title("Sales Distribution")
 plt.show()
+
 sns.histplot(df['Quantity'], kde=True)
 plt.title("Quantity Distribution")
 plt.show()
 
-7. Bivariate analysis
+# 7. Bivariate Analysis
 # Quantity vs Sales
 sns.scatterplot(x='Quantity', y='Sales', data=df)
 plt.show()
+
 # Discount vs Sales
 sns.scatterplot(x='Discount', y='Sales', data=df)
 plt.show()
+
 # Region vs Sales
 sns.boxplot(x='Region', y='Sales', data=df)
 plt.show()
 
-8. Multivariate analysis
+# 8. Multivariate Analysis
 sns.pairplot(df[['Quantity', 'Unit_Price', 'Discount', 'Sales']])
 plt.show()
 
-9. Detect sales outliers using IQR
-
+# 9. Detect Sales Outliers using IQR
 Q1 = df['Sales'].quantile(0.25)
 Q3 = df['Sales'].quantile(0.75)
 IQR = Q3 - Q1
 lower = Q1 - 1.5 * IQR
 upper = Q3 + 1.5 * IQR
+
 outliers = df[(df['Sales'] < lower) | (df['Sales'] > upper)]
 print("Number of outliers:", len(outliers))
 print(outliers)
 
-10. Handle outliers and 11. plot final distribution
+# 10 & 11. Handle Outliers and Plot Final Distribution
 df_clean = df[(df['Sales'] >= lower) & (df['Sales'] <= upper)]
 sns.histplot(df_clean['Sales'], kde=True)
 plt.title("Sales Distribution After Outlier Removal")
 plt.show()
-Conclusion: EDA helps understand sales distribution, relationships, regional patterns, missing values and unusual
-transactions. IQR is used to detect sales outliers.
-Question 2: Automated EDA and Statistical Relationship Analysis
-Aim: Analyze student academic performance and relationships between study habits, attendance, assignments and
-marks.
-A. Data Loading and Basic Analysis
+--------------------------------------------------------------------------------
+
+CONCLUSION:
+EDA helps understand sales distribution, relationships, regional patterns, 
+missing values, and unusual transactions. IQR is used to detect sales outliers.
+
+
+================================================================================
+Q2: AUTOMATED EDA AND STATISTICAL RELATIONSHIP ANALYSIS
+================================================================================
+
+AIM:
+Analyze student academic performance and relationships between study habits, 
+attendance, assignments, and marks using manual and automated EDA packages.
+
+PYTHON PROGRAM:
+--------------------------------------------------------------------------------
 import pandas as pd
 import numpy as np
 import matplotlib.pyplot as plt
 import seaborn as sns
+
+# A. Data Loading and Basic Analysis
 df = pd.read_csv("student_performance.csv")
 print(df.head())
 print(df.tail())
 print("Shape:", df.shape)
 print("Columns:", df.columns)
-print("Data Types:")
-print(df.dtypes)
-print("Statistical Summary:")
-print(df.describe())
-print("Missing values:")
-print(df.isnull().sum())
+print("Data Types:\n", df.dtypes)
+print("Statistical Summary:\n", df.describe())
+print("Missing values:\n", df.isnull().sum())
 print("Duplicates:", df.duplicated().sum())
-print("Unique values:")
-print(df.nunique())
-Handle missing and duplicate values:
+print("Unique values:\n", df.nunique())
+
+# Handle missing and duplicate values
 df = df.fillna(df.median(numeric_only=True))
 df = df.drop_duplicates()
-B1. D-Tale
-# Install once in terminal:
-# pip install dtale
+
+# B1. D-Tale
+# Terminal install: pip install dtale
 import dtale
 dtale.show(df)
-D-Tale provides an interactive overview, column statistics, missing values, correlations, histograms, scatter plots and
-outlier information.
-B2. YData Profiling / Pandas Profiling
-# Install once:
-# pip install ydata-profiling
+
+# B2. YData Profiling
+# Terminal install: pip install ydata-profiling
 from ydata_profiling import ProfileReport
 profile = ProfileReport(df, title="Student Performance Report")
 profile.to_file("student_profile.html")
-The HTML report contains dataset overview, variable statistics, missing values, correlations, distributions, duplicates and
-warnings.
-B3. Sweetviz
-# Install once:
-# pip install sweetviz
+
+# B3. Sweetviz
+# Terminal install: pip install sweetviz
 import sweetviz as sv
 report = sv.analyze(df)
 report.show_html("sweetviz_report.html")
-Sweetviz automatically shows distributions, associations, correlations, missing values and summary statistics.
-B4. AutoViz
-# Install once:
-# pip install autoviz
+
+# B4. AutoViz
+# Terminal install: pip install autoviz
 from autoviz.AutoViz_Class import AutoViz_Class
 AV = AutoViz_Class()
 AV.AutoViz("student_performance.csv")
-AutoViz automatically creates visualizations and helps identify important relationships.
-C. Covariance and Correlation
+
+# C. Covariance and Correlation Analysis
 num_cols = [
     'Study_Hours',
     'Attendance',
@@ -132,36 +146,53 @@ num_cols = [
     'Internal_Marks',
     'Final_Marks'
 ]
+
 covariance = df[num_cols].cov()
-print(covariance)
+print("Covariance Matrix:\n", covariance)
+
 correlation = df[num_cols].corr()
-print(correlation)
+print("Correlation Matrix:\n", correlation)
+
 plt.figure(figsize=(8, 6))
 sns.heatmap(correlation, annot=True, cmap='coolwarm')
 plt.title("Correlation Matrix")
 plt.show()
-Correlation ranges from -1 to +1. +1 means strong positive relationship, 0 means no linear relationship, and -1 means
-strong negative relationship. The strongest positive correlation with Final_Marks is the variable whose correlation value
-with Final_Marks is closest to +1. Weak variables have correlation values close to 0. Highly correlated independent
-variables have values close to +1 or -1.
-Conclusion: Automated EDA makes analysis faster by generating statistics, graphs, correlations, missing-value
-information and warnings.
-Question 3: Student Performance Prediction
-Aim: Predict Final_Marks using Linear Regression.
-Task 1. Load and preprocess
+--------------------------------------------------------------------------------
+
+CONCLUSION:
+Automated EDA tools (D-Tale, YData Profiling, Sweetviz, AutoViz) accelerate 
+analysis by generating statistics, plots, correlation matrices, missing-value 
+reports, and warnings automatically.
+
+
+================================================================================
+Q3: STUDENT PERFORMANCE PREDICTION (LINEAR REGRESSION)
+================================================================================
+
+AIM:
+Predict student Final_Marks using Multiple Linear Regression.
+
+PYTHON PROGRAM:
+--------------------------------------------------------------------------------
 import pandas as pd
 import numpy as np
+import matplotlib.pyplot as plt
+from sklearn.model_selection import train_test_split
+from sklearn.linear_model import LinearRegression
+from sklearn.metrics import r2_score, mean_squared_error
+
+# Task 1: Load and Preprocess Data
 df = pd.read_csv("student_performance.csv")
 print(df.head())
 print(df.shape)
 print(df.dtypes)
 print(df.describe())
 print(df.isnull().sum())
+
 df = df.fillna(df.median(numeric_only=True))
 df = df.drop_duplicates()
-Task 2. Apply Linear Regression
-from sklearn.model_selection import train_test_split
-from sklearn.linear_model import LinearRegression
+
+# Task 2: Apply Linear Regression
 X = df[
     [
         'Study_Hours',
@@ -172,116 +203,162 @@ X = df[
     ]
 ]
 y = df['Final_Marks']
+
 X_train, X_test, y_train, y_test = train_test_split(
     X, y, test_size=0.20, random_state=42
 )
+
 model = LinearRegression()
 model.fit(X_train, y_train)
+
 print("Intercept:", model.intercept_)
 for col, coef in zip(X.columns, model.coef_):
-    print(col, ":", coef)
+    print(f"{col}: {coef}")
+
 y_pred = model.predict(X_test)
-print("Predicted Final Marks:")
-print(y_pred)
-Task 3. Evaluate Linear Regression
-from sklearn.metrics import r2_score, mean_squared_error
+print("Predicted Final Marks:\n", y_pred)
+
+# Task 3: Evaluate Model
 r2 = r2_score(y_test, y_pred)
 mse = mean_squared_error(y_test, y_pred)
 rmse = np.sqrt(mse)
 residuals = y_test - y_pred
 RSS = np.sum(residuals ** 2)
-print("R2:", r2)
+
+print("R2 Score:", r2)
 print("MSE:", mse)
 print("RMSE:", rmse)
-print("Residuals:")
-print(residuals)
-print("RSS:", RSS)
+print("Residuals:\n", residuals)
+print("Residual Sum of Squares (RSS):", RSS)
+
 plt.scatter(y_test, y_pred)
 plt.xlabel("Actual Final Marks")
 plt.ylabel("Predicted Final Marks")
 plt.title("Actual vs Predicted Final Marks")
 plt.show()
-Interpretation: Higher R² generally means the model explains more variation in Final_Marks. Lower MSE and RMSE mean
-smaller prediction errors. Residual = Actual − Predicted.
-Question 4: Sales Data Analysis Using Python
-Aim: Load, clean, transform, analyze and visualize sales data.
-1. Load and inspect
+--------------------------------------------------------------------------------
+
+CONCLUSION:
+A higher R² indicates the model explains a significant portion of variance in 
+Final_Marks. Lower MSE and RMSE indicate smaller prediction error.
+
+
+================================================================================
+Q4: SALES DATA ANALYSIS AND VISUALIZATION USING PYTHON
+================================================================================
+
+AIM:
+Load, clean, transform, analyze, and visualize sales performance metrics.
+
+PYTHON PROGRAM:
+--------------------------------------------------------------------------------
 import pandas as pd
 import numpy as np
 import matplotlib.pyplot as plt
 import seaborn as sns
+
+# 1. Load and Inspect
 df = pd.read_csv("sales_data.csv")
 print(df.head())
 print(df.columns)
 print(df.dtypes)
 print(df.isnull().sum())
-2. Fill missing Price using average Price of its Product Category
+
+# 2. Fill Missing Price using Category Mean
 df['Price'] = df.groupby('Category')['Price'].transform(
     lambda x: x.fillna(x.mean())
 )
-print(df['Price'].isnull().sum())
-3. Create TotalAmount
+print("Missing Price values:", df['Price'].isnull().sum())
+
+# 3. Create TotalAmount Column
 df['TotalAmount'] = df['Quantity'] * df['Price']
 print(df.head())
-4. Data analysis
+
+# 4. Data Analysis
 # Highest total sales region
 region_sales = df.groupby('Region')['TotalAmount'].sum()
-print(region_sales)
+print("Region Sales:\n", region_sales)
 print("Highest Sales Region:", region_sales.idxmax())
+
 # Total Electronics revenue
 electronics_sales = df[
     df['Category'] == 'Electronics'
 ]['TotalAmount'].sum()
 print("Electronics Revenue:", electronics_sales)
+
 # Highest quantity in a single invoice
 highest = df.loc[df['Quantity'].idxmax()]
-print(highest)
+print("Invoice with Highest Quantity:\n", highest)
 
-5. Data visualization
-# Bar chart: sales by region
+# 5. Data Visualization
+# Bar Chart: Sales by Region
 region_sales.plot(kind='bar')
 plt.title("Total Sales by Region")
 plt.xlabel("Region")
 plt.ylabel("Sales")
 plt.show()
-# Pie chart: sales by category
+
+# Pie Chart: Sales by Category
 category_sales = df.groupby('Category')['TotalAmount'].sum()
 category_sales.plot(kind='pie', autopct='%1.1f%%')
 plt.title("Sales Distribution by Category")
 plt.ylabel("")
 plt.show()
-# Line chart: monthly sales
+
+# Line Chart: Monthly Sales
 df['Date'] = pd.to_datetime(df['Date'])
 monthly_sales = df.groupby(
     df['Date'].dt.to_period('M')
 )['TotalAmount'].sum()
 monthly_sales.plot(kind='line', marker='o')
-plt.title("Monthly Sales")
+plt.title("Monthly Sales Trend")
 plt.xlabel("Month")
 plt.ylabel("Sales")
 plt.show()
-Conclusion: The dataset was loaded, cleaned and transformed. Regional sales, electronics revenue and highest quantity
-were calculated, followed by bar, pie and line charts.
-Question 5: Employee Promotion Prediction Using Logistic Regression
-Aim: Predict promotion probability from years of work experience.
+--------------------------------------------------------------------------------
+
+CONCLUSION:
+The dataset was successfully cleaned and transformed. Key insights including 
+top regions, electronics revenue, and sales trends were identified via plots.
+
+
+================================================================================
+Q5: EMPLOYEE PROMOTION PREDICTION USING LOGISTIC REGRESSION
+================================================================================
+
+AIM:
+Predict promotion probability based on years of work experience using Logistic 
+Regression.
+
+PYTHON PROGRAM:
+--------------------------------------------------------------------------------
 import numpy as np
 import pandas as pd
 import matplotlib.pyplot as plt
 from sklearn.linear_model import LogisticRegression
-experience = np.array([1,2,3,4,5,6,7,8,9,10])
-promotion = np.array([0,0,0,0,0,1,1,1,1,1])
+
+# Create Sample Dataset
+experience = np.array([1, 2, 3, 4, 5, 6, 7, 8, 9, 10])
+promotion = np.array([0, 0, 0, 0, 0, 1, 1, 1, 1, 1])
+
 df = pd.DataFrame({
     'Experience': experience,
     'Promotion': promotion
 })
+
 X = experience.reshape(-1, 1)
 y = promotion
+
+# Train Logistic Regression Model
 model = LogisticRegression()
 model.fit(X, y)
-# Predicted probability
+
+# Predicted Probability
 probability = model.predict_proba(X)[:, 1]
-# Threshold 0.5
+
+# Decision Threshold 0.5
 predicted = (probability >= 0.5).astype(int)
+
 result = pd.DataFrame({
     'Experience': experience,
     'Actual': promotion,
@@ -289,39 +366,62 @@ result = pd.DataFrame({
     'Predicted': predicted
 })
 print(result)
-# Sigmoid curve
+
+# Plot Sigmoid Curve
 x_curve = np.linspace(1, 10, 100).reshape(-1, 1)
 y_curve = model.predict_proba(x_curve)[:, 1]
-plt.scatter(experience, promotion)
-plt.plot(x_curve, y_curve)
-plt.axhline(0.5, linestyle='--')
+
+plt.scatter(experience, promotion, color='red', label='Actual Data')
+plt.plot(x_curve, y_curve, color='blue', label='Logistic Curve')
+plt.axhline(0.5, linestyle='--', color='gray', label='Threshold (0.5)')
 plt.xlabel("Years of Experience")
 plt.ylabel("Probability of Promotion")
-plt.title("Logistic Regression - Promotion")
+plt.title("Logistic Regression - Promotion Prediction")
+plt.legend()
 plt.show()
-Interpretation: Logistic Regression gives a probability between 0 and 1. If probability ≥ 0.5, the employee is classified as
-Promoted (1); otherwise Not Promoted (0). As experience increases, the predicted promotion probability generally
-increases for this dataset.
-Question 6: Loan Approval Prediction Using Logistic Regression
-Aim: Predict loan approval probability from monthly income.
+--------------------------------------------------------------------------------
+
+CONCLUSION:
+Logistic Regression maps input features to probabilities between 0 and 1. 
+Probabilities >= 0.5 are classified as Promoted (1), otherwise Not Promoted (0).
+
+
+================================================================================
+Q6: LOAN APPROVAL PREDICTION USING LOGISTIC REGRESSION
+================================================================================
+
+AIM:
+Predict loan approval probability based on monthly income using Logistic Regression.
+
+PYTHON PROGRAM:
+--------------------------------------------------------------------------------
 import numpy as np
 import pandas as pd
 import matplotlib.pyplot as plt
 from sklearn.linear_model import LogisticRegression
-income = np.array([15,18,20,22,25,28,30,35,40,45])
-loan_approved = np.array([0,0,0,0,1,1,1,1,1,1])
+
+# Create Sample Dataset
+income = np.array([15, 18, 20, 22, 25, 28, 30, 35, 40, 45])
+loan_approved = np.array([0, 0, 0, 0, 1, 1, 1, 1, 1, 1])
+
 df = pd.DataFrame({
     'Income': income,
     'Loan_Approved': loan_approved
 })
+
 X = income.reshape(-1, 1)
 y = loan_approved
+
+# Train Logistic Regression Model
 model = LogisticRegression()
 model.fit(X, y)
-# Predicted probability
+
+# Predicted Probability
 probability = model.predict_proba(X)[:, 1]
-# Threshold 0.5
+
+# Decision Threshold 0.5
 prediction = (probability >= 0.5).astype(int)
+
 result = pd.DataFrame({
     'Income': income,
     'Actual': loan_approved,
@@ -329,19 +429,25 @@ result = pd.DataFrame({
     'Prediction': prediction
 })
 print(result)
-# Logistic curve
+
+# Plot Logistic Curve
 x_curve = np.linspace(15, 45, 100).reshape(-1, 1)
 y_curve = model.predict_proba(x_curve)[:, 1]
-plt.scatter(income, loan_approved)
-plt.plot(x_curve, y_curve)
-plt.axhline(0.5, linestyle='--')
+
+plt.scatter(income, loan_approved, color='red', label='Actual Data')
+plt.plot(x_curve, y_curve, color='blue', label='Logistic Curve')
+plt.axhline(0.5, linestyle='--', color='gray', label='Threshold (0.5)')
 plt.xlabel("Monthly Income")
 plt.ylabel("Probability of Loan Approval")
 plt.title("Logistic Regression - Loan Approval")
+plt.legend()
 plt.show()
-# Decision boundary
+
+# Decision Boundary Calculation
 boundary = -model.intercept_[0] / model.coef_[0][0]
-print("Decision Boundary:", boundary)
-Interpretation: The model converts income into a probability of loan approval. Probability ≥ 0.5 is classified as Loan
-Approved (1), while probability < 0.5 is classified as Loan Rejected (0).
-Quick Revision for Viva
+print("Decision Boundary (Income Threshold):", boundary)
+--------------------------------------------------------------------------------
+
+CONCLUSION:
+The model calculates a decision boundary for monthly income above which loans 
+are classified as Approved (1) and below which they are classified as Rejected (0).