INTERNSHIP REPORT

ON
AI/ML PROGRAMMING INTERNSHIP

Submitted in partial fulfillment of the requirement for the

award of the degree of

BACHELOR OF COMPUTER APPLICATION

Submitted by
Rasim Abdul Razak
U01FB22S0017
BCA-Vl SEM

Under the guidance of

Roch Libia Rani G
Assistant Professor

DE PAUL COLLEGE, MYSORE

DE PAUL COLLEGE, MYSORE -

571606

Belagola, Srirangapatna Taluk,Mandya district

Karnataka – 571 606
 CERTIFICATE

This is to certify that Industrial Training Internship Report entitled "AI/ML" has been
submitted by U01FB22S0017 Rasim Abdul Razak for partial fulfillment for the award
of the degree of Bachelor Of Computer Application affiliated to University of
Mysore during the year 2024-2025. It is certified that all the corrections/
suggestions indicated in the internal assessment have been incorporated in the
report and deposited in the library. The Internship has been approved as it satisfies
the academic requirements in respect of Internship work prescribed for the degree
in Bachelor of Computer Applications.

Roch Libia Rani

HOD
DE PAUL COLLEGE, MYSORE

Fr.Shijo Antony
Principal / Director
DE PAUL COLLEGE, MYSORE
 ATTENDANCE CERTIFICATE

This to certify that Rasim Abdul Razak programme at INTERNPE has

successfully completed an internship.

During this period, [he/she] demonstrated commendable dedication

and enthusiasm towards [his/her] assigned tasks and responsibilities.
[He/She] actively participated in various projects and initiatives,
contributing positively to the team's objectives.
Internship Supervisor: Krati Kumari

This certificate is awarded in recognition of [his/her] commitment and

valuable contribution to INTERNPE.

HR
INTERNPE
 INTERSHIP MENTOR DECLARATION

This is to certify that the Industrial Training Internship In

INTERNPE entitled AI/ML INTERNSHIP by Rasim Abduk Razak has been
done successfully and completed all the tasks provided in the internship.
 ACKNOWLEDGEMENT

I am deeply grateful to all those who contributed to the successful

completion of this Internship.
Firstly, I extend my thanks to the Almighty for His constant
guidance and support throughout this endeavor.
I express my sincere gratitude and thanks toRev. Dr. Baiju Antony CM,
Principal, De Paul College, for providing an opportunity to enrich my
knowledge.
I owe my sincere thanks toRev. Dr. Shijo Antony CM,Dean of Academics,
De Paul College, for providing the academic support.
I am immensely thankful toMrs.Roch Libia Rani, Head of the Department
of Computer Applications at De Paul College, for her invaluable assistance
and cooperation.
I pay my immense respect to my internal guideMrs. RochLibia Rani G,
Assistant Professor, Department of Computer Applications for the
unwavering support and valuable guidance throughout my internship
work.
I extend my appreciation to all the faculty and staff members of the BCA
department for their encouragement and assistance throughout this
journey.
I am indebted to my parents for their efforts and the opportunities they
have provided me. Their love and blessings have been instrumental in my
pursuit of knowledge and experience. Additionally, I am grateful to my
friends for their support in completing this Internship.
CHAPTER CONTENT PAGE NUMBER

I CHAPTER 1: 1

INTRODCUTION

II CHAPTER 2: 2

OBJECTIVE

III CHAPTER 3: 3-13

WEEKLY REPORT

IV CHAPTER 4: 14-15

SAMPLE CODING AND SCREENSHOTS

V CHAPTER 5: 16

CONCLUSION
 INTRODUCTION

This report presents my internship experience at INTERNPE, where I

worked on AI/ML-related projects. The internship provided me with an
opportunity to apply machine learning concepts to real-world problems
and develop practical solutions.
During the internship, I worked on several projects, including Diabetes
Prediction, Car Price Prediction, IPL Winning Team Prediction, Breast
Cancer Detection and Movie recommentation system . These projects
allowed me to explore various machine learning algorithms, data
preprocessing techniques, model evaluation methods, and deployment
strategies.
The experience helped me gain insights into the AI/ML domain, improve
my programming skills, and understand the challenges of working with
real datasets. Through this internship, I strengthened my knowledge of
Python, scikit-learn, TensorFlow, and data handling techniques.
 OBJECTIVES OF THE INTERNSHIP

The primary objectives of this internship were:

● To gain hands-on experience in AI/ML development by

working on real-world datasets and implementing machine
learning models.

● To enhance technical skills in Python programming, data

preprocessing, model training, and evaluation.

● To explore various machine learning techniques such as

classification, regression, and recommendation systems.

● To understand the importance of data preprocessing and

feature engineering in improving model performance.

● To deploy machine learning models and understand their

real-world applications.

● To develop problem-solving skills and learn how to apply

machine learning models effectively in different domains.
To collaborate and learn from experienced professionals in
the field of AI/ML.
 TASK DESCRIPTION
TASK 1:
Diabetes Prediction With ML:
Objective:

Data Preparation: Collect and preprocess relevant health data, including

features
like blood sugar levels and BMI.
Model Training: Train a machine learning model using a supervised
learning algorithm
on historical health data.
Prediction: Utilize the trained model to predict the likelihood of diabetes
based on
input features for early detection.
Accuracy Assessment: Evaluate the model's accuracy and performance
using
validation datasets to ensure reliable predictions.

How To Perform:

1. Data Collection:
-
Gather a comprehensive dataset containing relevant health information,
such as
blood sugar levels, BMI, age, and other potential features related to
diabetes.
2. Data Preprocessing:
-
Clean and preprocess the data to handle missing values, outliers, and
ensure it is
in a suitable format for machine learning algorithms.
-
Split the dataset into training and testing sets for model evaluation.
3. Feature Selection:
-
Identify and select the most relevant features that contribute to
predicting diabetes.
This may involve domain knowledge or using feature selection techniques.
4. Model Selection and Training:
-
Choose an appropriate machine learning algorithm for classification tasks,
such as
logistic regression, decision trees, or support vector machines.
-
Train the selected model on the training dataset, adjusting parameters for
optimal performance.
5. Model Evaluation:
-
Assess the model's performance on the testing dataset using metrics like
accuracy
precision, recall, and F1-score.
- Fine-tune the model if needed to improve its predictive
capabilities.
6. Prediction:
- Use the trained model to predict the likelihood of diabetes in
new, unseen data.
7. Interpretability (Optional):
- If applicable, explore methods for interpreting the model's
predictions, helping healthcare professionals understand the
factors influencing predictions.
8. Validation and Iteration:

Validate the model's predictions in real-world scenarios and

-

iterate on the model if necessary, incorporating feedback and new

data.
9. Deployment (Optional):

-If the model demonstrates high accuracy and reliability,

consider deploying it in a healthcare setting for real-time diabetes
risk assessment.
10. Ethical Considerations:

Be mindful of ethical considerations related to healthcare

-

data, ensuring privacy, security, and responsible use of predictive

models. 0

TASK 2:

CAR PRICE Predictor with ML TASK:

OBJECTIVE:

1. Data Collection: Gather a dataset containing features such as

car model, mileage, year, and price.
2. Model Training: Train a machine learning model, selecting an
algorithm suited for regression tasks.
3. Prediction: Enable the model to predict car prices based on
input features. 4. Accuracy Improvement (Optional): Optimize
the model's performance through techniques like feature
engineering, hyperparameter tuning, or using advanced algorithms.

HOW TO PERFORM:

Certainly! Here's a non-code, step-by-step guide on how to perform car

price prediction with machine learning: 1. Data Collection:

- Gather a diverse and comprehensive dataset containing

relevant features for car pricing, such as model, mileage, year, fuel
type, and other specifications.
2. Data Preprocessing:

-Clean and preprocess the dataset by handling missing values,

encoding categorical variables, and normalizing numerical features.
3. Model Selection:

- Choose a suitable machine learning regression model for car

price prediction, such as linear regression, decision trees, or
ensemble methods like random forests.
4. Training and Evaluation:

Split the dataset into training and testing sets.

-

- Train the selected model on the training set and evaluate its
performance on the testing set using metrics like Mean Squared
Error (MSE) or R-squared.
5. Hyperparameter Tuning (Optional):

-Optionally, fine-tune the model's hyperparameters to improve

its performance.
6. Prediction:
- Once the model is trained and evaluated, deploy it to make
predictions on new data, allowing users to input car features and
receive predicted prices.
7. Deployment (Optional):
 If applicable, deploy the trained model in a production
-

environment, such as a web application, to make real-time

predictions.
8. Monitoring and Maintenance (Optional):

Continuously monitor the model's performance over time and

-

update it as needed, considering changes in the car market or

additional features.
9. Documentation:

Provide documentation detailing the steps taken, the chosen

-

model, and any preprocessing techniques applied for future

reference or collaboration.

TASK 3:

IPL WINNING TEAM PREDICTION :

OBJECTIVE:

1. Data Collection: Gather historical data of Indian Premier

League (IPL) matches, including team performance, player
statistics, and match outcomes.
2. Model Development: Create a predictive model using machine
learning or statistical algorithms to analyze patterns and predict
winning teams.
3. Accuracy Evaluation: Assess the model's accuracy by
comparing its predictions with actual IPL match results.
4. Real-time Prediction: Implement the model to provide real-
time predictions for upcoming IPL matches based on current team
and player performance.

HOW TO PERFORM:

1. Data Collection:
- Gather historical data on Indian Premier League (IPL)
matches, including team performance, player statistics, and
match outcomes.
2. Feature Selection:
Identify relevant features that may influence a team's
-

performance and winning probability. This could include player

form, team rankings, past match results, and venue.
3. Data Preprocessing:

Clean and preprocess the collected data, handling missing

-

values, encoding categorical variables, and normalizing numerical

features.
4. Model Selection:

-Choose a suitable machine learning model for predicting

match outcomes. Common choices include classification algorithms
like logistic regression, decision trees, or ensemble methods.
5. Training the Model:

Split the dataset into training and testing sets.

-

- Train the selected model on historical data, using features to

predict the winning team.

6. Evaluation:
- Evaluate the model's performance using the testing dataset,
considering metrics like accuracy, precision, recall, or F1 score.
7. Feature Importance (Optional):

If applicable, analyze feature importance to understand which

-

factors significantly contribute to the prediction.

8. Model Deployment (Optional):

Optionally, deploy the trained model to a platform where it

-

can be used to make predictions for upcoming IPL matches.

9. Continuous Improvement:

-Regularly update the model with new data to improve its

accuracy and adapt to changing team dynamics and player
performances.
10. Consider External Factors:
 -Take into account external factors like player injuries, weather
conditions, and venue statistics that might impact match
outcomes.
11. User Interface (Optional):

Create a user-friendly interface for users to input relevant

-

data and receive predictions, if applicable. 12. Documentation:

Document the entire process, including data sources,

-

preprocessing steps, model selection, and evaluation metrics for

transparency and future reference.

TASK 4:

BREAST CANCER DETECTION TASK:

OBJECTIVE:

1.Feature Extraction: Identify relevant features from

mammography images for breast cancer detection.
2.Model Training: Train a machine learning model using labeled
data to recognize patterns indicative of cancerous tissues.
3.Prediction: Utilize the trained model to predict the likelihood of
breast cancer based on new mammogram images.
4.Accuracy Improvement: Continuously refine and optimize the
model to enhance accuracy and reduce false positives/negatives.

HOW TO PERFORM:

1. Data Collection:
- Collect a diverse dataset of mammography images, including
both cancerous and non-cancerous cases.
2. Data Preprocessing:

Preprocess the images, ensuring uniform size, resolution, and

-

standardization.
 Consider techniques like normalization and augmentation to
-

enhance dataset quality.

3.Feature Extraction:
- Identify relevant features from the preprocessed images, such as
texture, shape, and pixel intensity. 4. Model Selection:
Choose a suitable machine learning model (e.g., logistic
-

regression, support vector machines, or deep learning models like

CNNs) for classification.
5. Training:

Train the selected model using the labeled dataset, allowing the
-

algorithm to learn the patterns associated with cancerous and non-

cancerous tissues.
6. Validation:

Evaluate the model's performance on a separate validation

-

dataset to ensure it generalizes well to new data.

7. Hyperparameter Tuning (Optional):

If needed, fine-tune the model's hyperparameters to optimize its

-

performance.
8. Prediction:
- Utilize the trained model to predict the likelihood of breast
cancer in new mammogram images.
9. Evaluation:

- Assess the model's performance using metrics such as accuracy,

precision, recall, and F1-score.
10. Iterative Improvement:

-Continuously refine the model based on feedback, adding more

data or adjusting parameters to enhance its accuracy and reliability.
11. Deployment (Optional):

-If applicable, deploy the trained model in a clinical setting for

real-time breast cancer detection.
12. Ethical Considerations:
 - Consider ethical implications and biases associated with the
model's predictions, and ensure responsible deployment in
healthcare contexts.
PROJECT

MOVIE RECOMMENTATION SYSTEM with ML

OBJECTIVE:

The main objective of a movie recommendation system is to provide

personalized movie suggestions to users based on their preferences,
past interactions, and similarity to other users or movies. This
enhances user experience by reducing the effort quired to find
content that aligns with their tastes.

Types Of Recommendation System

● Content-Based Filtering:

Recommends movies similar to those the user has watched and

liked.
Uses features like genre, director, cast, etc.
Example: If a user likes action movies, it will recommend other
action movies.

● Collaborative Filtering:

User-based: Finds users with similar preferences and

recommends what those users liked.
Item-based: Finds movies that are often liked together and
recommends those.
Example: "Users who liked Movie A also liked Movie B".

Hybrid Models:
Combines content-based and collaborative filtering.
Example: Netflix uses a hybrid approach for better accuracy.
Popularity-Based:
Recommends movies based on overall popularity, irrespective of user
preferences.
Example: Trending movies or highest-rated movie.

Types of Recommendation Systems:

Step 1: Define the Objective

● What kind of recommendation system are you building (e.g.,

content based, collaborative)?
● Determine the dataset to use (e.g., user ratings, movie
metadata).

Step 2: Collect and Prepare Data

Dataset Sources:

● MovieLens Dataset: Popular dataset for movie

recommendations.
● Kaggle: Search for relevant movie datasets.

Data Cleaning:

● Handle missing values (e.g., incomplete ratings or metadata).

● Normalize data (e.g., scaling ratings).

Step 3: Exploratory Data Analysis (EDA)

● Analyze data distributions (e.g., most popular genres, user
activity).
 ● Visualize data using libraries like matplotlib or seaborn.

●
Step 4: Choose the Recommendation Algorithm

Content-Based:

● Use a similarity metric like Cosine Similarity.

● Use movie features (e.g., genre, keywords) and compute

similarity scores.

Collaborative Filtering:

● Implement using user-item interaction matrices.

● Use models like Matrix Factorization (e.g., Singular Value

Decomposition, SVD).

Step 5: Implement the System

Libraries and Tools:

● Python: pandas, numpy, scikit-learn

● Recommendation libraries: surprise, lightFM

from sklearn.metrics.pairwise import

cosine_similarity from
sklearn.feature_extraction.text import
CountVectorizer
# Example: Using genres for content-based recommendation
count = CountVectorizer(stop_words='english')
count_matrix =
count.fit_transform(movies['genres']) cosine_sim
= cosine_similarity(count_matrix, count_matrix)
# Function to get recommendations def
get_recommendations(title,
cosine_sim=cosine_sim):

idx = movies[movies['title'] == title].index[0]

sim_scores =
list(enumerate(cosine_sim[idx]))
sim_scores = sorted(sim_scores, key=lambda x: x[1],
reverse=True) sim_scores = sim_scores[1:11]
movie_indices = [i[0] for i in sim_scores] return
movies['title'].iloc[movie_indices]
print(get_recommendations("The Dark Knight"))

from surprise import SVD, Dataset, Reader

from surprise.model_selection import
train_test_split
# Load and prepare dataset
reader =
Reader(rating_scale=(1, 5))
data = Dataset.load_from_df(ratings[['userId', 'movieId', 'rating']],
reader) trainset, testset = train_test_split(data, test_size=0.25)
# Train model
algo = SVD()
algo.fit(trainset
)
# Predict ratings
predictions =
algo.test(testset)
print(predictions[:5])

Step 6: Evaluate the System

Metrics:
Root Mean Square Error (RMSE)
Mean Absolute Error (MAE)
Precision and Recall (for top-N recommendations)

from surprise import accuracy

accuracy.rmse(predictions) Step 7: Deploy the
System

Use a framework like Flask or Django to build a web

interface.
Integrate with a database to store user data and
interactions.
Deploy on a platform like Heroku or AWS.
SAMPLE SCREENSHOTS
 CONCLUSION

My AI/ML internship at INTERNPE has been an invaluable experience

that has significantly enhanced my technical skills and practical
knowledge in the field. Throughout the internship, I have had the
opportunity to work on diverse projects, collaborate with talented
professionals, and apply advanced machine learning algorithms to solve
real-world problems.
I have gained hands-on experience with data preprocessing, model
development, and deployment, which has strengthened my
understanding of the end-to-end machine learning pipeline. Additionally,
I have improved my proficiency in programming languages such as
Python and tools like TensorFlow, PyTorch, and scikit-learn.
One of the key highlights of this internship was the opportunity to
contribute to a significant project involving Breast Cancer
Prediction. This experience taught me the importance of teamwork,
communication, and continuous learning in a dynamic and fast-paced
environment.
Overall, this internship has been a stepping stone in my career, providing
me with the practical skills and confidence needed to pursue a successful
career in AI/ML. I am grateful for the mentorship and support from my
colleagues at INTERNPE and look forward to applying the knowledge and
experience gained in future endeavors.