Here is a complete, structured set of **sample programs** and practical file content suitable for

**Class 12 Artificial Intelligence (AI)** practical file (CBSE/equivalent curriculum), based on the
given topics.

### I. Python (Pandas) Programs

**Program 1: Create Pandas DataFrame from different sequence data types + basic display &
missing values**

```python

# Program 1: Create & Explore Pandas DataFrame

import pandas as pd

import numpy as np

# a) Create DataFrame using different sequence types

data = {

'Name': ['Amit', 'Priya', 'Rahul', 'Sneha', 'Vikram', 'Ananya', [Link], 'Rohan'],

'Age': [15, 16, 14, [Link], 17, 15, 16, 14],

'Marks': [85, 92, 78, 88, 95, [Link], 82, 90],

'Attendance': [95.5, 88.0, 92.5, 85.0, 98.0, 90.5, 87.0, 94.0]

df = [Link](data)

print("a) Complete DataFrame:")

print(df)

print("\n")
# b) First 5 records

print("b) First 5 records:")

print([Link]())

print("\n")

# c) Last 10 records (only 8 rows, so shows all)

print("c) Last 10 records:")

print([Link](10))

print("\n")

# d) Number of missing values in each column

print("d) Number of missing values:")

print([Link]().sum())

```

**Output (sample):**

```

a) Complete DataFrame:

Name Age Marks Attendance

0 Amit 15.0 85.0 95.5

1 Priya 16.0 92.0 88.0

...

d) Number of missing values:

Age 1

Marks 1

Attendance 0

dtype: int64

```

**Program 2: Download & Analyze CSV (Student performance / school data example)**

You can download a sample student dataset from Kaggle → "Predict Students' Dropout and
Academic Success"
([Link]
or any simple CSV with enrollment/attendance/dropout columns.

```python

# Program 2: Read CSV & perform statistical analysis

import pandas as pd

# a) Read CSV file (replace with your downloaded path)

df = pd.read_csv("student_dropout_data.csv") # e.g. downloaded file name

print("First 5 rows of dataset:")

print([Link]())

# Basic statistics

print("\nStatistical summary:")

print([Link]())
# b) Check missing values

print("\nMissing values in each column:")

print([Link]().sum())

# Additional analysis - correlation (example columns)

print("\nCorrelation between numeric columns:")

print(df.select_dtypes(include='number').corr())

# Example: Average dropout rate if column exists

if 'Dropout' in [Link]:

print("\nAverage dropout (1 = dropout):", df['Dropout'].mean())

```

**Paste in practical file:**

- Screenshot of code + output

- Screenshot of first 5 rows

- Screenshot of missing values & describe()

### II. Orange Data Mining Tool (Step-wise Procedures)

Orange is a free, open-source, no-code data mining tool (download from →

**1. Data Visualization in Orange (Step-wise)**

2. Connect File → Data Table (see data)

3. Connect File → Distributions (select any column e.g. Attendance → see histogram)

4. Connect File → Scatter Plot (select two features e.g. Marks vs Attendance)

5. Connect File → Heatmap (see correlations)

6. Connect File → Box Plot (compare marks across gender/category)

**Paste:** Screenshots of each widget + output plot

**2. Perform Classification in Orange**

Steps:

1. File → Load dataset (must have target/class column e.g. Dropout: Yes/No)

2. Connect File → Preprocess (optional – remove NaN, normalize)

3. Connect Preprocess → Random Forest / Tree / Naive Bayes (select learner)

4. Connect learner → Predictions (see predicted class)

5. Connect learner → Confusion Matrix (see accuracy, precision)

6. Connect learner → Test & Score (compare multiple models – 70:30 split, 10-fold CV)

**Paste:** Screenshot of workflow + Confusion Matrix + Test & Score results (accuracy e.g.
85–92%)

**3. Image Analytics / Classification in Orange**

Steps (needs Image Analytics add-on → Options → Add-ons → Install Image Analytics):

1. Import Images widget → load folder of images (e.g. two folders: Good Attendance vs Poor
Attendance photos/symbols)

2. Connect Import Images → Image Embedding (use Inception v3 or similar pretrained model)

3. Connect Image Embedding → Data Table (see numeric vectors)

4. Connect to Classification Tree / Random Forest

5. Connect to Confusion Matrix & Predictions

**Paste:** Screenshot of image grid + embedding + confusion matrix

**4. Word Cloud for Text Analysis in Orange**

Steps (needs Text Mining add-on):

1. Corpus widget → Load text file / paste student feedback texts

2. Connect Corpus → Preprocess Text (lowercase, remove stop words, stemming)

3. Connect Preprocess Text → Word Cloud

4. (Optional) Connect to Word Embedding → Clustering / Classification

**Paste:** Screenshot of beautiful Word Cloud (big words like "attendance", "meal", "dropout",
"motivation")

### III. Data Storytelling – Impact of Mid-Day Meal Scheme (MDMS) since 1995

**Key Findings from various studies (1995–2025):**

- MDMS launched nationally in **August 1995** as world's largest school feeding program

- Main goals → Improve **nutrition**, **enrollment**, **attendance** & reduce **dropout** rates
**Trends & Impact (from research & UDISE+ reports):**

- Enrollment increased significantly (especially girls) after 1995–2001 cooked meal phase

- Attendance improved by 5–15% in many states

- Dropout rates reduced remarkably (especially primary level) – many studies report impressive
impact on reducing dropout, especially among girls & disadvantaged groups

- Recent UDISE+ (2024–25): Primary dropout ~1.9%, but still higher in secondary level in states
like Bihar, UP

**Storytelling Narrative (Write in file):**

"The **Mid-Day Meal Scheme (MDMS)**, launched in 1995, has played a **pivotal role** in
transforming school education in India. By providing free nutritious meals, it removed hunger as
a barrier to schooling.

**Key patterns observed:**

- Sharp rise in enrollment & attendance post-1995 (especially girls)

- Significant reduction in dropout rates – studies show impressive improvement in retention

- Girls benefited more → helped reduce gender gap in education

**Visuals to include (Paste screenshots):**

- Line chart: Dropout rate trend (1995–2025) – showing downward trend after MDMS

- Bar chart: Attendance % before & after MDMS in sample states

- Word cloud: Keywords from student feedback about MDMS

**Conclusion:** MDMS is one of the most successful incentive-based policies in India that has
positively influenced student retention, especially in primary classes. However, challenges
remain in secondary level & private school coverage.
**Paste in file:**

- 4–5 screenshots of charts created in Orange/Pandas/Matplotlib

- Summary table of key statistics (e.g. Dropout before 1995: high → after: reduced)

All the above programs & steps are 100% practical-file ready for Class 12 AI. Take clear
screenshots of every step & output and arrange them neatly in your file. Good luck!