Information Processing, Networks and Analytics
Below are the InFoMM Mini-projects that fall within the mathematical area of information processing, networks and analytics. To read more about each project please click on the title, this will open the project's lay report:
- Bayesian networks for risk assessment (Leonardo)
- Cleansing of Financial Market Data: A Generic Framework (CME Group)
- Community structure in product-purchase networks (dunnhumby)
- Congestive heart failure triage prediction model (Iterex Therapeutics)
- Designing Subsampling Strategies (Whizz Education)
- Distribution of signal-to-interference ratio in wireless networks (BT)
- Improved source reconstruction from hydrophone data (NPL)
- Integrated information for organisational consciousness (BT)
- Matrix completion for drug discovery (e-Therapeutics)
- Matrix completion generalization to tensor completion and multi-way tables (Nielsen)
- Objective-Oriented Organisation Management (BT)
- Optimisation Under Uncertainty (dunnhumby)
- Pulse Clustering Using Probabilistic Methods (Thales)
- Short heuristics applied to purchasing decisions (Tesco)
- Source reconstruction from hydrophone data (NPL)
- Sparse Sensing matrices for compressed sensing and matrix completion (PA Consulting)
- Spatio-temporal networks in supermarkets (Tesco)
- Store Networks (dunnhumby)
- Understanding the Social Customer Using Topological Data Analysis (Emirates)
- User Cancellation Modelling: on Clustering of Customer Behaviours (Whizz Education)
- Classification of chaotic time series with deep learning (CCFE)
- Parameter Estimation in Agent-Based Models (Improbable)
- Uncertainty quantification for deep neural networks (NPL)
- Demand Transfer and Halo Effect on a Small Range of Products (Tesco)
- Bargaining Under Uncertainty (BP)
- Labour market impacts of the post-carbon transition in the US: A networks perspective (BEIS)
- Transfer Learning as a Technique to Utilise Machine Learning for Predictions of ITER Plasma Behaviour (CCFE)
- Sparse and Low-Rank Regularisation for Robust Deep Nets (NPL)