Multi- Class Adaptive Active Learning for Predicting Student Anxiety

1. INTRODUCTION

The Artificial Intelligence Internship conducted by ICT Academy in collaboration with Infosys
Foundation was an intensive, well-structured online training program aimed at bridging the gap
between academic knowledge and industry requirements in the field of Artificial Intelligence. With
the growing significance of AI across various sectors, this internship was tailored to equip
engineering students with the right set of technical skills and practical exposure required to stay
competitive in the modern digital workforce. As a final-year student pursuing Computer Science and
Engineering, this internship marked a significant milestone in my academic journey by introducing
me to state-of-the-art tools, industry practices, and the latest developments in AI and cloud
technologies.

The program was delivered via a blended learning approach, combining self-paced modules, live
virtual labs, and mentor-led sessions. It allowed participants to gain a comprehensive understanding
of AI concepts and their real-world applications. The curriculum was aligned with industry demands,
focusing not only on the theoretical aspects but also on hands-on experience with Microsoft Azure’s
AI services.

1.1 Relevance to Academic Learning

The internship directly complemented my academic curriculum by deepening my understanding of:

 Supervised and Unsupervised Learning techniques

 Data Preprocessing, feature selection, and model evaluation
 Natural Language Processing (NLP) and its applications
 Automation through Intelligent Document Processing

These topics enhanced my classroom knowledge and helped me connect theoretical concepts with
practical applications, providing a strong foundation for final-year project work and future
employment opportunities.

1.2 Program Structure

The internship was designed in a progressive format, starting from basic AI theory and gradually
moving to advanced topics and implementation:

 Phase 1: Introduction to Artificial Intelligence and Machine Learning

 Phase 2: Overview and usage of Microsoft Azure AI tools
 Phase 3: Building, training, and evaluating ML models using Azure ML Studio
 Phase 4: Natural Language Processing using LUIS
 Phase 5: Speech Translation and Cognitive Services
 Phase 6: Intelligent Document Processing using Form Recognizer

Each phase included assignments, lab demonstrations, and quizzes to ensure conceptual clarity and
hands-on proficiency.

Dept. of CS&E, MIT Mysore 2024-2025 1

Multi- Class Adaptive Active Learning for Predicting Student Anxiety

1.3 Learning Methodology

The internship followed an engaging and interactive learning methodology:

 Self-Learning Modules: Detailed video lectures and reading content

 Hands-On Labs: Real-time cloud-based practicals on Microsoft Azure
 Mentor Sessions: Weekly guidance and query resolution by industry experts
 Peer Collaboration: Encouragement to work with and learn from other interns
 Certification Assessments: End-of-module evaluations to test learning outcomes

This multi-faceted methodology helped in reinforcing learning, building confidence in using cloud
AI tools, and encouraging problem-solving approaches.

1.4 Skills and Competencies Gained

By the end of the internship, I had developed a strong foundational understanding of Artificial
Intelligence and Machine Learning concepts. The structured modules and guided sessions helped me
grasp essential theories such as supervised and unsupervised learning, model training, and evaluation
techniques. Moreover, through hands-on labs, I gained practical experience in working with
Microsoft Azure AI tools and services, which enabled me to build and deploy machine learning
models in a real-world cloud environment. These experiences enhanced my technical capabilities
and prepared me for implementing AI solutions beyond the academic setting.

Additionally, I acquired valuable insights into Natural Language Processing (NLP) workflows,
including how AI systems interpret, process, and generate human language. Working with tools such
as LUIS and Azure’s Speech Services allowed me to implement speech-to-text systems and
understand their real-world applications in customer service and accessibility. I also gained
proficiency in using AI for automating document handling tasks, which can greatly improve
efficiency in business environments. Alongside technical growth, the collaborative nature of the
program improved my communication, teamwork, and self-learning abilities, making me more
confident and industry-ready.

Dept. of CS&E, MIT Mysore 2024-2025 2

Multi- Class Adaptive Active Learning for Predicting Student Anxiety

2. COMPANY DETAILS

2.1 Overview of Organisation

ICT Academy, established through a public-private partnership model, is a not-for-profit society

supported by the Government, academia, and industry leaders. One of its key collaborators is the
Infosys Foundation, with whom it conducts industry-focused internship and training programs for
students across India. The organization’s primary objective is to bridge the gap between industry
expectations and the academic curriculum by equipping students with relevant skills and real-time
project experience. Through partnerships with global technology companies like Infosys, ICT
Academy delivers curated training programs that focus on emerging technologies such as Artificial
Intelligence, Cloud Computing, Data Science, and Cybersecurity. The Artificial Intelligence
Internship program is one such flagship initiative aimed at nurturing future-ready graduates with
hands-on skills and industry exposure.

ICT Academy operates with a clear focus on transforming higher education and making students
employable by integrating technology, training, and certification into the learning ecosystem. The
organization reaches institutions across urban and rural regions, providing equal learning
opportunities to all students, regardless of background. It plays a vital role in national skill
development initiatives and continues to be recognized for its innovative educational models and
effective collaboration with tech giants.

2.2 Mission and Vision

The mission of ICT Academy is to empower the youth of India through skill development programs,
creating a workforce that is ready to meet the challenges of the digital world. It seeks to prepare
students for next-generation jobs by offering cutting-edge, industry-relevant training in collaboration
with renowned technology companies and institutions.

The vision of ICT Academy is to become a globally recognized organization in transforming

academic institutions into talent development hubs by providing access to high-quality resources,
world-class mentorship, and certifications. The organization aims to bridge the gap between
academia and industry by creating a knowledge ecosystem that supports inclusive growth, digital
literacy, and innovation in education.

2.3 Services and Offering

ICT Academy offers a wide range of services that target capacity building and employment
readiness among students and educators. Its offerings include:

 Industry-led Internships and Certification Programs in domains like Artificial

Intelligence, Machine Learning, Data Science, Cybersecurity, Cloud Computing, and Full
Stack Development.

Dept. of CS&E, MIT Mysore 2024-2025 3

Multi- Class Adaptive Active Learning for Predicting Student Anxiety

 Faculty Development Programs (FDPs) that enhance the teaching capabilities of educators
in technical and non-technical areas.
 Institutional Membership Services that provide colleges and universities with access to
digital learning platforms, student assessments, and curriculum support.
 Research and Innovation Support, where students and faculty are encouraged to work on
real-world problems and contribute to applied research projects.
 Career and Employability Enhancement Programs, such as job fairs, career counseling,
and corporate readiness training.

In collaboration with the Infosys Foundation, ICT Academy launched the Artificial Intelligence
Internship Program to expose students to the real-world application of AI tools and techniques.
This initiative focused on Microsoft Azure, allowing students to build and deploy AI models,
thereby enhancing their technical proficiency and employability in today’s competitive job market.

Dept. of CS&E, MIT Mysore 2024-2025 4

Multi- Class Adaptive Active Learning for Predicting Student Anxiety

3. ARTIFICIAL INTELLIGENCE

3.1 Introduction

Artificial Intelligence (AI) is a branch of computer science focused on building systems capable of
performing tasks that typically require human intelligence. These tasks include learning, reasoning,
problem-solving, understanding natural language, and perceiving speech and images. The goal of AI
is to create machines that can mimic cognitive functions, adapt to new inputs, and improve over time
through learning. AI has become one of the most transformative technologies in the modern digital
era, driving innovation across industries such as healthcare, finance, education, transportation, and
manufacturing.

During the internship offered by ICT Academy in collaboration with Infosys Foundation, we were
introduced to various AI subdomains, including Machine Learning (ML), Natural Language
Processing (NLP), Speech Recognition, and Intelligent Document Processing. The program
emphasized not only theoretical understanding but also practical application using Microsoft Azure’s
AI services. Through hands-on labs, we learned how to train models, deploy AI solutions in the
cloud, and use them in real-world scenarios like translating speech, extracting data from documents,
and understanding human language inputs. The internship provided a complete learning experience
by integrating technology, use cases, and development best practices, offering a strong foundation in
AI fundamentals.

The internship also highlighted the ethical considerations and challenges associated with AI
development. We learned about the importance of data privacy, fairness, and transparency in AI
models. This included understanding biases in data and algorithms that can lead to unintended
discrimination or errors. By incorporating ethical AI principles, we were encouraged to design
systems that are not only powerful but also responsible and trustworthy. This perspective is crucial
as AI technologies become more integrated into everyday life and critical decision-making
processes.

Another valuable aspect of the program was exposure to various AI tools and frameworks beyond
Azure, such as TensorFlow, PyTorch, and open-source NLP libraries. This broadened our
understanding of the AI ecosystem and helped us appreciate the diversity of approaches for building
intelligent applications. The hands-on projects pushed us to experiment with data preprocessing,
feature engineering, model tuning, and evaluation techniques, thereby deepening our practical skills.
Collaborative teamwork and peer discussions further enriched the learning experience by allowing
us to troubleshoot challenges and share innovative ideas.

Finally, the internship prepared us for the future of work in an AI-driven world by emphasizing
continuous learning and adaptability. We explored emerging trends like AutoML, conversational AI,
and edge computing, which are reshaping how AI is developed and deployed. The program also
encouraged building a portfolio of AI projects and certifications to demonstrate our competencies to
potential employers. Overall, this immersive experience not only strengthened our technical

Dept. of CS&E, MIT Mysore 2024-2025 5

Multi- Class Adaptive Active Learning for Predicting Student Anxiety

capabilities but also inspired us to contribute meaningfully to the evolving field of artificial
intelligence.

3.2 Fundamental Needs

To effectively implement and work with Artificial Intelligence, certain fundamental elements are
essential, both in terms of technical infrastructure and conceptual understanding. One of the most
critical needs is data—AI systems rely on vast amounts of structured and unstructured data to learn
patterns and make accurate decisions. The quality and volume of data directly impact model
performance. Hence, effective data collection, cleaning, and preprocessing are foundational steps in
building any AI solution.

In addition to data, computational resources are a core requirement. AI models, especially those
involving deep learning, need high-performance computing environments, which are provided
efficiently through cloud platforms like Microsoft Azure. Tools such as Azure Machine Learning
Studio allow developers to build, train, test, and deploy models in a scalable and cost-effective
manner. Alongside infrastructure, there is a need for skilled professionals who understand
algorithms, model evaluation, API integration, and ethical considerations in AI. Lastly, industry-
focused training and guided mentorship—such as that offered through this internship—are vital to
bridge the gap between academic theory and industry application, preparing students to contribute
effectively to the AI ecosystem.

4. PROJECT DETAILS

The practical component of the internship involved executing various AI-based mini-projects on the
Azure platform. The first project focused on creating a regression model using AutoML, which
involved data cleaning, selecting features, training the model, and interpreting the results using
different evaluation metrics. Following this, we implemented a language understanding model using
Azure LUIS, where we defined intents, provided sample utterances, and trained the model to
respond intelligently to user inputs. The third project involved speecg detection, where we used
Azure’s cognitive services to convert spoken language into text and translate it into multiple
languages, showcasing the power of AI in cross-linguistic communication. Finally, we worked on
Intelligent Document Processing, using Azure Form Recognizer to extract structured data from
unstructured forms such as receipts or invoices. These projects helped in reinforcing the theoretical
concepts learned during sessions and provided hands-on experience in deploying AI models. Each
project mimicked real-world scenarios where AI can replace manual processes, improve accuracy,
and save time and resources.

Dept. of CS&E, MIT Mysore 2024-2025 6