MILESTONE ACADEMY
(Session 2025-26)
ARTIFICIAL INTELLIGENCE (843)
PROJECT TOPIC :
CAPSTONE PROJECT:
DATA ANALYSIS USING ORANGE
DATA MINING TOOL
Submitted in partial fulfillment of AISSCE 2025-26
Curriculum for the Subject Artificial Intelligence (843)
GUIDED BY : SUBMITTED BY :
MRS. ANJU VERMA ADHYA BHAGAT
MILESTONE ACADEMY CLASS : 12 th
1
MILESTONE ACADEMY
CERTIFICATE
This is to certify that ADHYA BHAGAT
student of class XII has successfully
completed her project on
CAPSTONE PROJECT:
DATA ANALYSIS USING ORANGE DATA
MINING TOOL
Under the guidance of subject teachers
during the session 2025–26 in partial
fulfillment of the Artificial Intelligence (843)
practical examination conducted by
AISSCE.
PRINCIPAL
EXTERNAL EXAMINER INTERNAL EXAMINER
2
ACKNOWLEDGEMENT
I would like to express my sincere gratitude to
all those who guided and supported me in
completing this Artificial Intelligence
(Subject Code 843) Capstone Project.
I am deeply thankful to my subject teachers for
their constant guidance, valuable insights, and
encouragement throughout this project.
My heartfelt thanks to our respected principal,
Madam, for fostering a motivating environment
for learning and innovation.
I am also grateful to my parents for their
continuous support, understanding, and
encouragement, and to my friends for their
cooperation and helpful suggestions.
Thank you all for your guidance and
contribution in making this project a successful
learning experience.
Adhya Bhagat
Class : 12 th B
3
PREFACE
This Artificial Intelligence (Subject Code 843)
project has been prepared as part of the Class
XII CBSE curriculum. It represents an effort to
apply classroom learning to practical and
analytical tasks using data-driven approaches.
The project aims to develop critical thinking,
problem-solving, and technical skills while
exploring real-world applications of Artificial
Intelligence. I hope this work provides valuable
insights and contributes to a better
understanding of AI concepts.
4
INTRODUCTION
Orange Data Mining
Orange Data Mining is an open-source software suite for
datavisualization, machine learning, and data mining. It
operates onavisual programming paradigm, allowing users to
build data analysis workflowsby connecting widgets on a
canvas.
Key aspects of Orange Data Mining:
Visual Programming :
Users drag and drop ‘widgets’ onto a canvas and connect them
with ‘lines’ to define data flow and analysis steps. These lines
represent the flow of data and signals between different
processing units.
5
Widgets :
These are the fundamental building blocks of Orange
workflows, each performing a specific data operation.
Examples include widgets for loading (e.g., File), preprocessing
(e.g., Select Columns), visualization (e.g., Scatter Plot),
modeling (e.g., Logistic Regression), and evaluation (e.g., Test &
Score).
Lines:
The lines connecting widgets on the canvas represent the flow
of data or signals. When a line is drawn from one widget to
another, it signifies that the output of the first widget will be
used as the input for the second. This visual connection
establishes the order and dependencies within the data
analysis pipeline.
Component-Based Approach :
Orange's design allows for modularity, where complex analyses
arebroken down into smaller, manageable components
(widgets) that canbe easily combined and rearranged.
User-Friendly Interface :
The graphical user interface (GUI) makes Orange accessible
touserswith varying levels of programming experience,
frombeginners toexperienced data scientists.
Applications :
Orange is widely used for exploratory data analysis, data
visualization, building and evaluating machine learning models,
and educational purposes in data science and machine
learning.
6
TOPIC
“SENTIMENTAL ANALYSIS USING ORANGE DATAMINING”
The Text Mining add-on installed
To install the Text add-on :
1. Open Orange.
2. Go to Options > Add-ons.
3. Search for Text, then install it.
4. Restart Orange
Steps for Sentiment Analysis in Orange :
1. Load the Data
Use the "File" widget to import your dataset (CSV, Excel, etc.)
containing text.
➡ Ensure the text column (like "Review") is properly labeled in
your data.
Preprocess the Text :
Add the "Corpus" widget and connect it to File.
Then, connect Corpus to "Preprocess Text" to clean and
normalizethetext.
7
In Preprocess Text :
1. Tokenize text
2. Lowercase
3. Remove stop words
4. Lemmatize
Add Sentiment Lexicon (Optional for Lexicon-Based) :
If you want to use a lexicon-based method:
Use the "Sentiment Analysis" widget.
It uses a built-in sentiment lexicon (e.g., Vader).
Train a Model (for ML-Based Sentiment Analysis) :
If your data includes sentiment labels (e.g., Positive/Negative),
you can train a classifier:
Workflow Overview
File → Corpus → Corpus Viewer → Preprocess Text Sentiment
→ Analysis → Select columns → Heat map → Corpus Viewer →
Word cloud → Data table → Bar P
8
CAPSTONE
Purpose:
Show understanding of AI concepts (like data, models,
neural networks, etc.)
Practice problem-solving and critical thinking.
Learn how to collect and prepare data
Design and train a basic AI model (using tools like
Teachable Machine, Google Colab, or Scratch, depending
on your level) Present your work in a structured format
9
Components of the AI Capstone Project :
1. Problem Identification
Choose a real-world problem you want to solve.
Example: "Detecting if someone is wearing a mask or not"
2. Data Collection
Gather data relevant to your problem.
This can be images, text, or numbers.
You may use tools like Google Forms, Kaggle, or collect your
own data.
3. Data Exploration and Analysis
-Visualize and understand the data.
-Use tools like Excel, Python (pandas/matplotlib), or just charts.
4. Model Building
Build a simple AI model
Use platforms like:
Teachable Machine (for image, audio, pose recognition)
Scratch + ML Extensions (block-based)
Python (Colab) if you are more advanced
5. Model Testing and Evaluation
-Test your model’s accuracy
-Identify if it gives correct or incorrect outputs
6. Presentation and Documentation
-Create a project report or presentation.
-Include introduction, methodology, data, results, conclusion.
10
The 5 Stages of the Capstone Project Cycle :
1. Problem Scoping :
Goal: Understand and define the real-world problem you want
to solve.
Key Steps:
Identify a problem in society, school, home, or environment.
Understand why the problem matters and who it affects.
Define the goal of your AI project.
Example: "People often forget to wear masks in public during
health emergencies. This can be a health risk. We want to build
a model to detect if a person is wearing a mask or not."
2. Data Acquisition :
Goal: Collect the right data that your AI model will learn from.
Key Steps:
Decide what type of data you need (images, numbers, text,
audio).
Collect or download data from reliable sources. Organize the
data into categories (labels).
TOOLS:
1. Google Forms
2. Teachable Machine
3. Kaggle datasets
4. Collect your own data (photos, etc.)
Example: "Collect 500 images: 250 of people wearing masks
and 250 without masks."
11
3. Data Exploration :
Goal: Understand, clean, and prepare the data for model
training
Key Steps:
Check if the data is balanced and good quality.
Clean the data (remove errors or duplicates).
Visualize the data using charts/graphs.
TOOLS:
1. Excel/Google Sheets
2. Python (optional)
3. Charts and graphs
Example: "Make a pie chart showing 50% images with masks
and 50%without."
4. Modelling :
Goal: Train an AI model using your prepared data.
Key Steps:
Choose the right AI tool.
Train your model using labeled data.
Test the model with new (unseen) data.
TOOLS:
1. Teachable Machine (best for image/audio/pose)
2. Scratch with AI extension
3. Python (for advanced students)
Example: Train a model in Teachable Machine with the two
image classes: "Mask"and "No Mask."
12
5. Evaluation :
Goal: Test how accurate and useful your AI model is.
Key Steps:
Check if the model gives correct predictions.
Find the accuracy percentage.
Note down any errors or improvements needed.
Example: Test 10 new images. If the model correctly identifies 9
out of 10, the accuracy is 90%.
13
14
15
16
17
18
19
20
21
22
23
24
25