Scribd
Upload
3 views2 pages

Python Assignment

The document outlines a Python assignment for MCS-405, which includes tasks related to data analysis using Pandas, NumPy, and machine learning techniques for real estate price prediction and retail customer segmentation. Key tasks involve data cleaning, exploratory data analysis, modeling, and clustering, with specific datasets provided for each focus area. Expected deliverables include Jupyter Notebooks and Markdown reports detailing the analysis and insights derived from the datasets.

Uploaded by

triedsomethingnew6
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
3 views2 pages

Python Assignment

The document outlines a Python assignment for MCS-405, which includes tasks related to data analysis using Pandas, NumPy, and machine learning techniques for real estate price prediction and retail customer segmentation. Key tasks involve data cleaning, exploratory data analysis, modeling, and clustering, with specific datasets provided for each focus area. Expected deliverables include Jupyter Notebooks and Markdown reports detailing the analysis and insights derived from the datasets.

Uploaded by

triedsomethingnew6
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Python assignment

MCS – 405

1. Using Pandas, load a CSV dataset and compute mean, median, and standard
deviation. Explain why these measures are important.
2. Write a Python program to clean a dataset (handle missing values and
duplicates).Why is data cleaning necessary before analysis?
3. Perform basic EDA using Matplotlib or [Link] histogram and boxplot.
What insights can you derive?
4. Explain [Link] a Python program to generate a correlation matrix
and interpret results.
5. Using NumPy, implement mean and variance [Link] with built-
in functions.

6. Real Estate Price Prediction (Regression Focus)


Dataset: California Housing Prices (Kaggle)
Problem Statement: A real estate investment firm wants to build an automated tool to
predict the median house value in any given district in California based on local
demographics and geographical data.

Tasks:
i. GitHub Setup: Create a new repository and initialize a [Link].
ii. Data Operations: Load the dataset using Pandas. Handle missing values.
iii. Exploratory Data Analysis (EDA): Plot a histogram for median_house_value.
Create a correlation heatmap to see which features (like median_income) correlate
most strongly with house price.
iv. Modeling: Split the data into Training and Testing sets. Train on these models
using Scikit-Learn.
 Linear Regression
 Ridge/Lasso
 XGBoost

v. Evaluation: Calculate the error of the model's predictions. Choose which one
performed best?
 RMSE, MAE

Expected Deliverable: A Jupyter Notebook with commented code and a


Markdown conclusion explaining which feature impacts house prices the most.
Learning Outcomes
 Regression techniques
 Feature engineering
 Real-world dataset handling

7. Retail Customer Segmentation (Unsupervised Learning Focus)


Dataset: E-Commerce Data / Online Retail (Kaggle)

Problem Statement: An online retail store wants to launch targeted marketing


campaigns. Since they have no labels for "good" or "bad" customers, they need a data
scientist to find hidden groups (clusters) of similar buyers based on their purchasing
history.
Tasks:
i. Heavy Data Cleaning: Remove cancelled orders (invoice numbers starting with
'C') and drop rows where the CustomerID is missing (Null).
ii. Feature Engineering (RFM): Use [Link]() to calculate three new features
for each customer:
a. Recency: Days since last purchase.
b. Frequency: Total number of purchases.
c. Monetary: Total money spent.
iii. Clustering: Apply the K-Means Clustering algorithm to group customers into 3 to
5 distinct segments.
iv. Data Visualization (2D/3D plots)
v. Business Insight:
 High-value customers
 Target groups
Expected Deliverable: A Markdown report outlining the specific marketing
strategy the business should use for each of the discovered customer clusters.

Learning Outcomes
 Unsupervised learning
 Data visualization

You might also like