Delhi Public School, Etawah

In the fulfilment of Class XII

A Project on

“Flower Species Prediction”

Submitted By Submitted To
Mridul Dixit. Mrs. Randeep Kaur
 CERTIFICATE
This is to certify that MRIDUL DIXIT of class 12th has
successfully completed the project in chemistry on topic
“FLOWER SPECIES PREDICTION TOOL” in the
fulfilment of Class XII for the Subject ARTIFICIAL
INTELLIGENCE (843) during the academic year 2025-
26.
Subject In-charge External Principal
 ACKNOWLEDGEMENT

Primarily I would Thank GOD, THAT WHO IS for being able to

complete this project with success. Then I would like to thank my
Parents who helped me with their valuable suggestions.

Then I would like to thank our Principal Ma'am Mrs Bhavana

Singh and My AI Teacher whose guidance has been the ones that
helped me patch my project with success.

Last but not the least I would thank my friends and classmates
who helped me a lot.

MRIDUL DIXIT
XII (A)
INTRODUCTION TO ORANGE TOOL
Orange is an open-source machine learning and data visualization tool. It uses a simple drag-and-drop
workflow with widgets to perform tasks such as:
• Data importing
• Data visualization
• Classification
• Clustering
• Model training
• Model evaluation

Advantages of Orange:
• No coding required
• Visual and beginner-friendly
• Instant graphs and model outputs
• Supports Machine Learning
OBJECTIVE OF THE PROJECT
• To understand machine learning classification using Orange
• To import and explore the Iris dataset
• To visualize dataset using graphs
• To create a Decision Tree model
• To evaluate the model using Test & Score
• To understand accuracy and confusion matrix

TOOLS AND TECHNOLOGIES USED

Tool Purpose
Orange Tool Machine Learning, Visualization
Iris Dataset Classification Dataset
Decision Tree Algorithm Model Building
Test & Score Widget Model Evaluation
DATASET DESCRIPTION – IRIS DATASET
The Iris dataset contains 150 flower samples with 5 attributes:
Features
1. Sepal Length
2. Sepal Width
3. Petal Length
4. Petal Width
5. Species (Target Class)
o Iris Setosa

o Iris Versicolor

o Iris Virginica

It is one of the most commonly used datasets for learning classification.

AI PROJECT CYCLE STEPS
The AI Project Cycle consists of five steps

STEP 1: PROBLEM SCOPING

Problem Statement:
Can we classify a flower species based on sepal and petal measurements?
Stakeholders:
Students, botanists, ML learners.
Success Criteria:
High model accuracy and clear class separation.
.

STEP 2: DATA ACQUISITION

The dataset used is the Iris dataset, which is preloaded in Orange.
It contains structured numerical data suitable for classification.
Source: Open-source dataset available within Orange.
STEP 3: DATA EXPLORATION
Tools used in Orange:
• Data Table → View dataset
• Scatter Plot → Visualize feature relationships
• Box Plot → Study data distribution
• Feature Importance → Identify useful features
Key Observations:
• Petal length and width clearly separate Setosa from others.
• Versicolor and Virginica slightly overlap.

STEP 4: MODELLING
A Decision Tree Classifier is used for model building.
Workflow:
1. File → Import dataset
2. Scatter Plot → Visualize classes
3. Decision Tree → Build classification model
4. Test & Score → Evaluate
5. Confusion Matrix → Analyze predictions
STEP 5: EVALUATION
Model evaluation is done using:
• Accuracy
• Precision & Recall
• Confusion Matrix
• Cross Validation
Result:
The Decision Tree achieved around 95% accuracy, showing it is effective for this dataset.

WORKING / METHODOLOGY
Step-by-step procedure in Orange:
1. Open Orange Tool
2. Drag “File” widget → Load [Link]
3. Connect to “Data Table” to view data
4. Connect to "Scatter Plot" → observe class separation
5. Add “Decision Tree” widget → train model
6. Add “Test & Score” → evaluate model
7. Add “Confusion Matrix” → analyze predictions
WORKFLOW DIAGRAM
File → Data Table → Scatter Plot → Decision Tree → Test & Score → Confusion Matrix

RESULTS & FINDINGS

• Scatter plot shows clear separation between species
• Decision Tree model achieved high accuracy (~95%)
• Most predictions are correct
• Minor confusion occurs between Versicolor and Virginica
• Orange tool simplified ML implementation
CONCLUSION
The project demonstrates the working of a machine learning model using Orange.
The Iris dataset was successfully classified using a Decision Tree model.
This project helped in understanding data visualization, model building, accuracy evaluation, and practical
Machine Learning concepts without using programming.

BIBLIOGRAPHY
• Orange Data Mining Tool
• CBSE Artificial Intelligence Study Material
• Iris Dataset Documentation
• Wikipedia & Kaggle (Dataset Explanation)
• Class 12 AI Textbook