Orange Data Mining Class 10 AI: Penguins, Coral, Sentiment

Start here

Courses behind these walkthroughs

The Orange course where every widget below is built and explained, the 417 course whose batch runs these use cases inside the full syllabus, and the Python course for students who want to see the same models in code.

Visual Machine Learning with Orange for Teens course thumbnail

The tool / Orange

Visual ML with Orange

File to Predictions on tables, images and text, with time to try other learners and settings until the readouts make sense.

Open the course →

CBSE Computational Thinking and AI for Teens course thumbnail

The batch / 417 Class 10

CBSE AI for Class 9 to 12

The full 417 course, where each Orange walkthrough sits in the unit it belongs to and is rehearsed before the practical.

Open the course →

Same models in code / Python

Python for Teens

Pandas, NumPy and Matplotlib for Unit 7, and the code version of the steps Orange hides behind its widgets.

Open the course →

Ans. The short version

Orange data mining in Class 10 AI is the no-code tool CBSE 417 uses for three use case walkthroughs in 2026-27: predicting Palmer penguin species from body measurements (Unit 4, Statistical Data), classifying coral photos as bleached or unbleached (Unit 5, Computer Vision) and scoring the sentiment of text (Unit 6, NLP). Each follows the AI project cycle with widgets such as File, Impute, Data Sampler, Test and Score, Confusion Matrix and Predictions. These units are assessed in the practical. Modern Age Coders builds them live in a Monday and Wednesday 9:00 PM IST batch, or one to one.

Q1. Which Orange use cases are in the syllabus?

Which no-code AI use cases does CBSE name for Class 10 in 2026-27?

The 2026-27 CBSE 417 curriculum names three Orange walkthroughs, one per practical domain, and the handbook adds a short price-prediction example. All of them are assessed through the practical, not the theory paper.

Use case	Unit and domain	Model type	Orange add-on needed
Palmer penguins: predict the species	Unit 4, Statistical Data	Classification, three classes	None, core Orange
Food price index: predict the price (handbook example)	Unit 4, Statistical Data	Regression	None, core Orange
Coral bleaching: bleached or unbleached?	Unit 5, Computer Vision	Image classification, two classes	Image Analytics
Sentiment analysis of reviews or posts	Unit 6, NLP	Positive, negative or neutral scoring	Text

Before the first class: install and add-ons

Install Orange
Download it from orangedatamining.com and install it like any desktop program.
Add the two add-ons
Open Options, then Add-ons, tick Image Analytics and Text, and restart Orange.
Check the toolbox
The left panel should now show Image Analytics and Text sections beside Data, Visualize, Model and Evaluate.

A missing add-on is the commonest reason a student cannot find Import Images or Sentiment Analysis on the day.

Teachable Machine and Lobe

The same curriculum names two more no-code tools for Unit 5: Teachable Machine and Lobe, for a "smart sorter" that tells bottles, cans and paper apart. They train an image model in the browser or an app from photos you take. Orange differs because every step of the project cycle is a visible widget you can open and explain, which is why the walkthroughs below use it.

Where these units sit in the 50 practical marks is on CBSE Class 10 AI practical exam questions.

Q2. How do I build the Palmer penguins model?

Palmer penguins class 10 AI: the Orange workflow, step by step

The penguins model predicts a penguin's species, Adelie, Chinstrap or Gentoo, from its island and four body measurements. CBSE's own workflow file uses eleven widgets, and each one maps to a stage of the AI project cycle.

The data. CBSE's training file has 344 penguins: 152 Adelie, 124 Gentoo and 68 Chinstrap. Columns are species, island, culmen length and culmen depth (the culmen is the upper ridge of the bill), flipper length and body mass. Two penguins have all four measurements missing, which is why the workflow needs Impute.

Step	Widget	Connect	What to set or read	Project cycle stage
1	File (renamed TrainData)	Starts the chain	Load penguins_train.csv; check species shows as categorical	Data acquisition
2	Impute	TrainData to Impute	Fill the missing measurements, for example with the average	Data exploration
3	Feature Statistics	Impute to Feature Statistics	Read mean, spread and missing count per column; mean culmen length is about 43.9 mm	Data exploration
4	Select Columns	Impute to Select Columns	Move species to Target; keep island and the four measurements as features	Modelling (set up)
5	Data Sampler	Select Columns to Data Sampler	Fixed proportion 80 percent, stratified; about 275 penguins train, 69 are held back	Modelling (split)
6	Data Info, twice	Data Sample to one, Remaining Data to the other	Confirm the two row counts add up to 344	Checking the split
7	Tree and Random Forest	Data Sample to each learner	Defaults are fine to start	Modelling
8	Test and Score	Data Sample to Data, Remaining Data to Test Data, both learners to Learner	Choose "Test on test data"; compare CA, Precision, Recall and F1 for the two models	Evaluation
9	File (renamed TestData)	Starts a second chain	Load penguins_test.csv	New data
10	Predictions	TestData to Data; Tree and Random Forest Model outputs to Predictors	Each row shows the true species beside each model's guess	Deployment, in miniature

What the exploration should show

Island gives a lot away. Gentoo appear only on Biscoe and Chinstrap only on Dream; Adelie live on all three islands.
Flippers separate Gentoo. Their average flipper is about 217 mm against about 190 mm for Adelie.
Culmen length separates Adelie from Chinstrap. About 38.8 mm against about 48.8 mm on average.

Drop a Scatter Plot on the Impute output with culmen length against flipper length, coloured by species, and the three groups are visible before any model is trained.

Reading the Confusion Matrixillustrative, 69 test penguins

                 Predicted
Actual      Adelie Chinstrap Gentoo
Adelie        29       1        0
Chinstrap      2      12        0
Gentoo         0       0       25

Correct: 29 + 12 + 25 = 66 of 69
Accuracy: 66 / 69 = 0.957
Chinstrap recall: 12 / 14 = 0.857
Adelie precision: 29 / 31 = 0.935

Your numbers will differ with each random split. The pattern is typical: Gentoo is never confused, while Adelie and Chinstrap swap a few, because their bodies are similar and Chinstrap has the fewest examples.

One honest caution about the CBSE test file

CBSE's penguins_test.csv holds the same 344 penguins as the training file, with the species column renamed "actual species". So the Predictions table will look almost perfect: the models have already seen those birds. The fair score is the one Test and Score reports on the 69 held-back penguins. If an examiner asks why Predictions looks better than Test and Score, that is the answer, and it is the same idea as never testing on training data in Unit 3. The full evaluation reasoning is on CBSE Class 10 AI evaluating models questions.

Q3. How does Orange classify coral photos?

Coral bleaching project class 10: the image classification chain in Orange

Orange cannot learn from raw pixels directly. The Image Embedding widget turns every photo into a long row of numbers that describe its features, and from then on an ordinary classifier treats the photos like rows in a table.

The data. Two folders, Train and Test. Inside each, one sub-folder per class: Bleached and Unbleached. Import Images reads the sub-folder name as the label, so the folder layout is the labelling.

Step	Widget	Connect	What to set or read
1	Import Images (renamed Train Data)	Starts the chain	Choose the training folder; it should report two categories
2	Image Viewer	Train Data to Image Viewer	Scroll the photos; check for blurred, duplicate or wrongly filed images
3	Image Embedding	Train Data to Image Embedding	SqueezeNet runs offline and is the default; Inception v3 and the others need internet
4	Data Table	Image Embedding to Data Table	See each image now as a row of numeric columns plus its category
5	Logistic Regression, Random Forest, SVM	Each learner to Test and Score	Three models, so their scores can be compared
6	Test and Score	Image Embedding to Data	Cross-validation is a sound choice here; compare CA, F1, Precision and Recall
7	Confusion Matrix	Test and Score to Confusion Matrix	Switch between learners on its left to see which one misses bleached coral
8	Import Images (Test Data), Image Viewer, Image Embedding	A second chain for the test folder	Point it at the test folder, not the training one
9	Logistic Regression (second copy)	Train embeddings to its Data	Trains the chosen model on all training photos
10	Predictions	Test embeddings to Data; the model to Predictors	Each test photo with its predicted class and probability

Which model won

In the CBSE walkthrough, Logistic Regression gave the best accuracy of the three, which is why the final chain uses it. On your own photos another learner may win; the method of choosing is what earns marks.

Which error matters

A bleached reef predicted as unbleached is a false negative, and it is the costly one, because the reef gets no attention. So read recall for the Bleached class, not only overall accuracy.

A slip in the steps PDF

CBSE's step sheet tells you, at the prediction stage, to select the training folder. Choose the test folder: predicting photos the model trained on proves nothing.

Pixels, RGB channels and convolution, the theory behind what Image Embedding does, are on CBSE Class 10 AI computer vision questions. Coral as a full project with its 4Ws is on CBSE Class 10 AI project ideas.

Q4. How is sentiment analysis done in Orange?

Sentiment analysis in Orange for the NLP unit

Sentiment analysis decides whether a piece of text is positive, negative or neutral. In Orange, the Text add-on loads the texts as a corpus, cleans them, and the Sentiment Analysis widget adds a score to every document.

The curriculum names a sentiment walkthrough for Unit 6, but the shared CBSE folder for it does not hold a workflow file we could open, so the chain below is the standard route the Text add-on documents. Schools may use their own review or tweet dataset.

Step	Widget	What it does	What to set or read
1	Corpus (or Import Documents)	Loads the texts as documents	Pick the column that holds the text as the text feature
2	Preprocess Text	Text normalisation	Lowercase, tokenise, remove punctuation and stopwords: the Unit 6 theory, as tick boxes
3	Sentiment Analysis	Scores each document	Choose Vader: it adds positive, negative, neutral and compound columns; compound runs from -1 to +1
4	Data Table	Shows the scores	Sort by compound to see the most negative and most positive texts
5	Word Cloud	Shows frequent words	Compare clouds for the positive and the negative texts
6	Box Plot or Distributions	Summarises the scores	How many texts fall on each side of zero

Five sentences, VADER compound scorerun in Python

"The new library is bright,
 quiet and wonderful."           0.765
"The bus was late again and
 the driver was rude."          -0.459
"The canteen opens at noon."     0.000
"The food was not bad at all."   0.431
"Great, another power cut during
 my online exam."                0.459

Scored with the VADER lexicon, the method behind Orange's Vader option; Orange may differ in the last decimal. The last line is sarcasm, and the tool calls it positive.

What the five scores teach

Neutral is real. A plain fact scores 0, neither positive nor negative.
Negation is handled. "Not bad" comes out mildly positive, because VADER reads "not" before "bad".
Sarcasm is not. "Great" pulls the power-cut complaint positive. A lexicon counts words; it does not understand intent.

That last point answers a favourite viva question on the limits of NLP. Bag of Words and TFIDF, the counting ideas underneath, are worked by hand on Bag of Words and TFIDF for Class 10 AI.

Q5. What is no-code AI, and what will I be asked?

No code AI class 10: price prediction, code versus no-code, and viva questions

No-code AI means building, training and testing a model by dragging and connecting ready-made blocks instead of writing a program. Low-code tools mix blocks with a little code.

The handbook's regression example

In Unit 4 the handbook tells of Kayla, who budgets animal food at a zoo and needs to predict rising food prices without knowing how to code. The chain is short:

File
Load the food price index data and set the price index as the target.
Data Table
Look at the rows before modelling.
Linear Regression
File to Linear Regression; the output is a number, so this is regression.
Test and Score
File to Data, Linear Regression to Learner; read RMSE, MAE and R2 instead of accuracy.
Predictions
Data to Data, and the Linear Regression model to Predictors.

The handbook's last step joins Test and Score to Predictions and mentions a logistic model. In Orange, Predictions takes data plus a trained model, as in the last step above, and a price is predicted with linear, not logistic, regression.

Code versus no-code for statistical data

Point	Code (Python)	No-code (Orange)
Skill needed	Programming	Understanding the steps
Flexibility	Anything you can write	Only the widgets that exist
Speed to a first model	Slower	Minutes
Risk	Bugs in your code	Trusting outputs without checking them, called automation bias

The handbook also warns that no-code platforms give less control over security, so they suit data that is not sensitive.

Questions examiners ask at the Orange screen

Why did you use Select Columns?
Answer
To mark species as the target, the label the model must predict. Without a target, Test and Score has nothing to check against.
What does Data Sampler do, and why stratified?
Answer
It splits the data into a training part and a held-back test part. Stratified keeps the same share of each species in both parts.
What does Image Embedding output?
Answer
A row of numbers per image that describes its features, so a normal classifier can learn from photos.
Which stage of the AI project cycle is Test and Score?
Answer
Evaluation. Impute and the plots belong to data exploration; the learners to modelling.
Is your sentiment tool always right?
Answer
No. It scores word lists, so sarcasm and mixed reviews fool it. Checking a sample by hand shows how often it agrees with a person.

More of these, for every practical unit, are on CBSE Class 10 AI viva questions.

Q6. How is Orange taught live?

How the live batch teaches Orange for the 417 practical

Following a step sheet builds a workflow once. Explaining every widget, and rebuilding it on data you have not seen, is what the practical and its viva reward, so that is what class time goes on.

Built on a shared screen

The teacher builds one use case live while students build alongside on their own machines. Each widget is opened, its settings changed and the effect on the readout talked through.

Same widgets, new data

The following class swaps the dataset, fruit photos for coral, shop reviews for tweets, and students rebuild the chain without the step sheet. The gaps show quickly and are fixed in the same session.

Explain it aloud

Every student explains one widget per class in the words an examiner expects: target, held-back data, false negative, embedding. That rehearsal is what turns a working canvas into viva marks.

Students who are stuck on an add-on install or a broken connection share their screen and it is fixed live. How these walkthroughs feed the project and file is on the project and practical file page, and the project cycle they all follow is on AI project cycle and ethics.

Batch or personalised: you choose

Build these Orange workflows in the Monday and Wednesday batch, or one to one

In the batch, students build each use case alongside classmates and hear each other explain the widgets. One to one goes at your child's speed on their own laptop, with every install problem and broken link fixed on the spot. Same syllabus, same teachers, monthly billing, no admission fee, stop at the end of any month.

Option 1 · Fixed batch timings

Group batch, Monday and Wednesday 9 PM

₹1,499

per month, two live classes a week

Monday and Wednesday, 9:00 PM IST, live on video
Fifteen to twenty Class 10 AI students, one teacher
One use case built live each week
The same chain rebuilt on unseen data
Widget explanations rehearsed for the viva

Book a free demo in this batch

Option 2 · Choose your own timing

Personalised one to one

₹4,999

per month, one private class a week, at a time set with you

Days and time chosen by you, in your own time zone
Orange set up and checked on your own computer
Extra datasets for the use case your child finds hardest
The school's own practical tasks practised if you share them
Same CBSE 417 syllabus and teachers

Ask for a personalised plan

A student whose school lab time is short, or who has never installed Orange, often moves fastest one to one, where setup and each workflow are done on their own machine. A student who also needs the theory units gains more from the batch, where Orange sits inside the full 417 course. Fees are shown in your own currency.

Studying outside India?

Orange and no-code AI practice for CBSE students in the Gulf and abroad

CBSE schools abroad set the same practical units, and Orange runs the same on any laptop. Here is the evening batch in your local time this winter.

Where you live	Mon and Wed 9:00 PM IST	Best format
India (IST) every Indian city	9:00 PM	Batch fits
UAE and Oman Dubai, Abu Dhabi, Sharjah, Muscat	7:30 PM	Batch fits
Saudi Arabia, Qatar, Kuwait, Bahrain Riyadh, Jeddah, Dammam, Doha, Kuwait City, Manama	6:30 PM	Batch fits
Nepal Kathmandu	9:15 PM	Batch fits
Ethiopia and East Africa Addis Ababa and East Africa Time	6:30 PM	Batch fits
Singapore and Malaysia Singapore, Kuala Lumpur	11:30 PM	Choose 1-on-1

If the batch hour clashes with school or travel, personalised one to one runs at a time you choose. More for families outside India is on CBSE Class 10 AI online classes in the Gulf and abroad.

What families say

Rated 4.9 across 547 Google reviews

Real reviews from real families. We neither write nor commission them.

★★★★★

"The one step solution for my son. Modern Age Coders make learning coding so simple that kids love it. The teachers explain complex concepts clearly with practical exercises and interactive content."

Ria Mukherjee

Parent

★★★★★

"Modern Age Coders has been a game-changer for me. I struggled to grasp IT concepts and coding before joining, but their classes transformed everything. I can now confidently write complex programs with ease."

Samriddha Mondal

Student

★★★★★

"One of the most wonderful education centres out there. Education is not limited to school syllabus but focuses on skill development."

Vansh Agarwal

Student

★★★★★

"My child Dhairya is really enjoying the Modern Age Coders classes. This is his first online class and he eagerly looks forward to it. I can already see his improvement, and the teachers are very cooperative."

Sonam Oswal

Parent of Dhairya

★★★★★

"Modern Age Coders have wonderful teachers who teach in a clear, easy and practical way. The teacher boosts students' confidence and inspires them to learn without hesitation."

Sonu Goyal

Parent

★★★★★

"I highly recommend this computer coding class! The teachers are incredibly knowledgeable and passionate about coding."

Ritu Kedia

Parent

The rest of the Class 10 series

More pages for Class 10 board students

Every page in the CBSE Class 10 AI (417) series, grouped by what you need: classes, each unit, question practice, notes and revision, and the exam itself.

Classes, tuition and batches

Unit by unit

Questions and practice

Notes and revision

Exam, marks and practicals

Choosing the subject

ICSE Class 10 Computer Applications (Java): The Tue and Wed ICSE Java batch ICSE Computer Applications syllabus ICSE Java programs practice ICSE CA important questions

Questions about Orange in Class 10 AI

What students ask about Orange Data Mining in class 10 AI

What is Orange Data Mining in class 10 AI?

Orange is a free, open-source tool for data analysis and machine learning in which you build a model by dragging widgets onto a canvas and connecting them, without writing code. CBSE 417 uses it in 2026-27 for three practical walkthroughs: Palmer penguins (Statistical Data), coral bleaching (Computer Vision) and sentiment analysis (NLP).

Is Orange Data Mining in the class 10 AI theory exam?

The walkthroughs belong to Units 4, 5 and 6 practical work, which is assessed through the practical file, practical exam and viva. Units 4 and 7 are practical only. Objective items about no-code tools can still appear, so know what Orange is and what its main widgets do.

What is the Palmer penguins case study in class 10 AI?

It is the Unit 4 use case in which an Orange model predicts a penguin's species, Adelie, Chinstrap or Gentoo, from its island, culmen length and depth, flipper length and body mass. CBSE's training file has 344 penguins. The workflow runs File, Impute, Select Columns, Data Sampler, Tree and Random Forest, Test and Score, then Predictions.

Which widgets are used for the coral bleaching project?

Import Images to load photos from Bleached and Unbleached sub-folders, Image Viewer to check them, Image Embedding to turn each photo into numbers, then Logistic Regression, Random Forest and SVM compared in Test and Score with a Confusion Matrix. A second Import Images chain feeds the test photos into Predictions.

Why does Image Embedding need the internet?

Only some embedders do. SqueezeNet, the default in current versions, runs on your own computer. Inception v3, VGG-16, VGG-19 and the others send images to a server to be embedded, so they need a connection. If the lab has no internet, choose SqueezeNet.

How do I do sentiment analysis in Orange?

Install the Text add-on, load your texts with Corpus, clean them with Preprocess Text, then connect Sentiment Analysis and choose a method such as Vader. It adds positive, negative, neutral and compound scores. View them in a Data Table sorted by compound, and use Word Cloud to compare frequent words.

What is no code AI in class 10?

No-code AI means building and testing a model through a visual interface, dragging and connecting blocks, instead of programming. Examples in the curriculum are Orange, Teachable Machine and Lobe. Low-code tools combine blocks with some code. The trade-off is speed and ease against flexibility and control.

Is Teachable Machine in the class 10 AI syllabus?

Yes. The 2026-27 CBSE 417 curriculum lists Teachable Machine with supervised learning in Unit 2 and, with Lobe, for a smart sorter activity in the Unit 5 practical, where a model learns to tell bottles, cans and paper apart from photos.

Why are my Predictions almost 100 percent correct?

Usually because you are predicting rows the model was trained on. CBSE's penguins test file repeats the training penguins, so it will look near perfect. The honest measure is Test and Score on the data held back by Data Sampler.

Is Orange practice part of your regular classes?

Yes. Each use case is built live and then rebuilt on fresh data in the batch or one to one, all covered by the monthly fee shown on this page in your own currency. There is no admission fee and the first class is a free demo.

Start

Build a free Orange workflow with a teacher

In the free first class your child builds the penguins workflow with the teacher, from File to Predictions, and explains what three of the widgets do. You will see how the practical is prepared, and the mentor will tell you plainly which parts of the practical need work and whether the batch or one to one suits your child.

Want to see how the practical units fit the full subject first? Read the 417 syllabus explained.

WhatsApp us · +91 91233 66161 · connect@modernagecoders.com

Every class is live video. The form requests a callback and reserves nothing.

Want it sooner? Book a priority demo

A full live class of about 45 to 60 minutes at a slot you pick, with a written skill report and learning roadmap on WhatsApp afterwards. The fee is adjusted against your first month if you enrol.

Book a priority demo

Courses behind these walkthroughs

Visual ML with Orange

CBSE AI for Class 9 to 12

Python for Teens

Which no-code AI use cases does CBSE name for Class 10 in 2026-27?

Before the first class: install and add-ons

Install Orange

Add the two add-ons

Check the toolbox

Teachable Machine and Lobe

Palmer penguins class 10 AI: the Orange workflow, step by step

What the exploration should show

One honest caution about the CBSE test file

Coral bleaching project class 10: the image classification chain in Orange

Which model won

Which error matters

A slip in the steps PDF

Sentiment analysis in Orange for the NLP unit

What the five scores teach

No code AI class 10: price prediction, code versus no-code, and viva questions

The handbook's regression example

File

Data Table

Linear Regression

Test and Score

Predictions

Code versus no-code for statistical data

Questions examiners ask at the Orange screen

How the live batch teaches Orange for the 417 practical

Built on a shared screen

Same widgets, new data

Explain it aloud

Build these Orange workflows in the Monday and Wednesday batch, or one to one

Group batch, Monday and Wednesday 9 PM

Personalised one to one

Orange and no-code AI practice for CBSE students in the Gulf and abroad

Rated 4.9 across 547 Google reviews

More pages for Class 10 board students

Classes, tuition and batches

Unit by unit

Questions and practice

Notes and revision

Exam, marks and practicals

Choosing the subject

What students ask about Orange Data Mining in class 10 AI

What is Orange Data Mining in class 10 AI?

Is Orange Data Mining in the class 10 AI theory exam?

What is the Palmer penguins case study in class 10 AI?

Which widgets are used for the coral bleaching project?

Why does Image Embedding need the internet?

How do I do sentiment analysis in Orange?

What is no code AI in class 10?

Is Teachable Machine in the class 10 AI syllabus?

Why are my Predictions almost 100 percent correct?

Is Orange practice part of your regular classes?

Build a free Orange workflow with a teacher

Want it sooner? Book a priority demo

Keep exploring Modern Age Coders

By class and age

Learn more

Free resources

From the blog

Start here