COMP 5112
Algorithms for Data Science
Algorithmic techniques to handle (massive/big) data arising from, for example, social media, mobile devices, sensors financial transactions. Algorithmic techniques may include locality-sensitive hashing, dimensionality reduction, streaming, clustering, VC-dimensions, external memory, core sets, link analysis and recommendation systems.
2026
Topics
- Topics may include
- Approximation algorithms design techniques
- Dimensionality Reduction
- Online Algorithms (including the role of Primal-Dual LPs in their analysis)
- Finding Similar Items using Locality-Sensitive Hashing
- Nearest Neighbor Searching
- Clustering
- FPT Algorithmic Design Techniques
- Algorithmic Aspects of Social Networks (Graph Partitioning, Searching various substruc
- tures
- Multiplicative-Weight Update Method
- amount of lecture time available.
- Mining of Massive Datasets (MMDS) by Leskovec, Rajaraman, Ullman
- Some Journal and Conference Articles.
- For project ideas, look for conferences such as SODA, WWW, KDD, Data Minining
Sources
- Course catalogue ↗2026–2027
- Algorithms for Data Science ↗2026