DATA
ANALYTICS
Our Mission
Intern Vision is committed to empowering interns through
visionary guidance and practical experience.
Our mission is to equip aspiring professionals with the tools,
skills, and mindset needed to succeed in the evolving tech
landscape. We strive to create a platform where innovation meets
opportunity, shaping confident and capable future leaders.
About Internship
COMPLETION PLACEMENT NETWORK
CERTIFICATE SUPPORT OPPORTUNITY
DATA ANALYTICS
TASK LIST
You will need to complete at least 2 tasks for
successful completion of internship.
TASK 1
Exploratory Data Analysis (EDA) on
a Public Dataset
Goal: Perform exploratory data analysis on a real-world dataset and generate insights using visualizations
and summary statistics.
Tech Stack (optional): Python · Pandas · Matplotlib or Seaborn · Jupyter Notebook
Requirements:
Choose a public dataset (Kaggle, UCI, etc.)
Clean and preprocess the data (handle missing values, etc.)
Generate visualizations (histograms, boxplots, heatmaps, etc.)
Calculate and explain descriptive statistics
Write conclusions and insights in markdown cells
Docs (optional):
Pandas · Seaborn · Kaggle Datasets
TASK 2
Build an Interactive Dashboard Using Plotly and
Dash
Goal: Create a web-based interactive dashboard to visualize and filter data dynamically.
Tech Stack (optional): Python · Pandas · Plotly · Dash · Jupyter or standalone app
Requirements:
Use a real or cleaned dataset
Visualize key metrics using graphs (bar, line, pie, etc.)
Add dropdowns, sliders, and date pickers to filter data
Make layout responsive and organized
Write a brief explanation of insights shown on the dashboard
Docs (optional):
Dash Docs · Plotly Express
TASK 3
Predictive Modeling with Machine Learning
(Classification Problem)
Goal: Build a classification model using a real-world dataset to predict a target variable, evaluate its
performance, and interpret results.
Tech Stack (optional): Python · Pandas · Scikit-learn · Matplotlib or Seaborn · Jupyter Notebook
[Optional] XGBoost or LightGBM for advanced learners
Requirements:
Choose a public classification dataset (e.g., Titanic, Heart Disease, etc.)
Clean and preprocess the data (handle missing values, encode categories, scale if needed)
Split dataset into training and test sets
Train at least two classification models (e.g., Logistic Regression, Decision Tree, Random Forest)
Evaluate performance using accuracy, precision, recall, F1-score, and confusion matrix
Use feature importance to interpret model behavior
Visualize key results and comparisons
Docs:
Scikit-learn Classification · Train/Test Split · Confusion Matrix
TASK 4
SQL Data Analysis on a Relational Database
Goal: Write and execute SQL queries to analyze data from a relational database and extract actionable insights.
Tech Stack (optional): PostgreSQL or MySQL · pgAdmin / DBeaver / SQLiteStudio · [Optional] Google
BigQuery or DB Fiddle
Requirements:
Use a sample database (e.g., Chinook, Northwind)
Write SQL queries to answer at least 8–10 analytical questions such as:
What are the top-selling products?
Which customers generate the most revenue?
What is the monthly trend of orders?
Use aggregate functions (SUM, AVG, COUNT), joins, subqueries, and GROUP BY, ORDER BY clauses
Document each query’s purpose and result as comments or markdown
Docs:
SQL Tutorial · SQL Aggregate Functions · PostgreSQL Docs
TASK 5
A/B Test Analysis and Business Recommendation
Goal: Analyze A/B test results to determine if a new feature (e.g., new UI, pricing model, or campaign) led
to statistically significant improvement.
Tech Stack (optional): Python · Pandas · SciPy or Statsmodels · Matplotlib / Seaborn · Jupyter Notebook
Requirements:
Use a public or mock dataset (e.g., Kaggle: A/B Testing)
Define hypotheses (null and alternative)
Perform statistical significance testing (e.g., two-sample t-test or
z-test)
Calculate and interpret p-values and confidence intervals
Visualize results (e.g., conversion rates, distributions)
Provide a clear business recommendation based on findings
Docs:
AB Testing with Python · Scipy t-test · Statsmodels
CONNECT
WITH US
@InternVision
[Link]
CineMovie App
Chess I/M
[Link]
[Link]@[Link]
Intern Vision