Study Jam: Introduction to Machine Learning

With the primary objective of introducing students to the fundamentals of Machine Learning (ML) and providing them with hands-on experience in building models, the Study Jam on Introduction to Machine Learning was conducted during 06-07  August 2025 by Tinkerhub. The program was designed to cater to both beginners and enthusiasts, ensuring that participants could grasp the concepts regardless of their prior exposure to AI or data science.

The sessions were conducted online from 7:30 PM to 8:30 PM, allowing attendees from diverse locations and backgrounds to join conveniently without geographical limitations. The event was hosted by Nandith Narayanan (S3, CSE Dept), who ensured smooth coordination and active engagement throughout. The technical sessions were led by an experienced professional, Mr Joel Basil Kurian (AI Mentor), who brought both clarity and industry insights to the discussions. His teaching style blended theory with practical applications, making the content both approachable and relevant.

Approximately 40–50 participants attended the Study Jam, comprising students, hobbyists, and individuals eager to understand the basics of Machine Learning. The diverse backgrounds of participants enriched the discussions, as questions and perspectives from different domains were actively shared.

Day 1 – Introduction to Core Concepts

On the first day, participants were introduced to fundamental ML concepts such as:

• Supervised and Unsupervised Learning

• Dataset structure, features, and labels

• Overfitting and Underfitting

• Model Evaluation

The session also provided a detailed exploration of the Naive Bayes algorithm, covering:

• Its mathematical foundation, including Bayes’ theorem

• The application of probability in classification problems

• Real-world applications (spam detection, sentiment analysis, etc.)

• Advantages and limitations compared to other algorithms

Day 2 – Practical Session on House Price Prediction

The second day was a dedicated guided hands-on session focusing on a regression problem: House Price Prediction. Participants learned the complete workflow, which included:

1. Dataset Handling – Importing and understanding dataset structure

2. Data Preprocessing – Handling missing values, encoding categorical variables, and feature scaling

3. Model Training – Selecting and fitting a regression model

4. Evaluation Metrics – Using Mean Absolute Error (MAE), Mean Squared Error (MSE), and R² Score to measure performance

5. Result Interpretation – Understanding how predictions can be applied in real-world housing markets

Live coding demonstrations were used to bridge the gap between theory and practice. Participants could follow along, replicate the steps, and experiment with modifications in real time.

The event stood out for its interactive and learner-centered approach. The host facilitated Q&A sessions after each major concept, ensuring that doubts were clarified immediately. The speaker encouraged participants to think critically about model performance and data quality, thereby planting the seed for deeper exploration into advanced ML techniques such as decision trees, neural networks, and ensemble learning in future sessions.

The feedback collected after the sessions was overwhelmingly positive. Participants appreciated the balance between conceptual clarity and hands-on practice. Many expressed interest in extended workshops on advanced topics like deep learning, natural language processing, and deploying ML models in production. Several attendees also noted that the session motivated them to initiate their own small projects, including stock price prediction, customer segmentation, and chatbot development.

The Editorial Team of  News & Events joins the entire Vidya fraternity in extending heartiest congratulations to Goutham on the successful conduct  of the session that strengthens participants’ foundational knowledge sparks genuine curiosity and enthusiasm for AI and data science !!!