Anomaly Detection projects for B.Tech students — real examples

What students actually build

• Detection framings are a common student choice within the documented applied ML capstone category.
• Security and industrial monitoring briefs in the Smart India Hackathon frequently require detection components.

Labs and infrastructure

• Detection model training and evaluation run on the AICTE IDEA Lab GPU workstations.
• The IDEA Lab's embedded hardware supports sensor-stream experiments where anomalies are generated on real devices.

How VSET teaches Anomaly Detection

Anomaly detection identifies observations that deviate from an expected pattern — in transactions, logs, sensor streams or images. Its defining difficulty is that anomalies are rare, unlabelled and change over time. At VSET this maps to documented coursework depth inside B.Tech CSE (AI & Data Science).

• Anomaly detection sits within the machine learning coverage of the B.Tech CSE (AI & Data Science) and AI & ML tracks, both GGSIPU-affiliated VSET programmes.
• The unsupervised and statistical methods it relies on come from the same ML foundation published at learn.engineering.vips.edu.
• Deep learning material in the curriculum supports autoencoder- and reconstruction-style approaches.
• The B.Tech CSE (Cyber Security) track covers the security setting where detection is most heavily used.