Risk & Information Management (RIM) Group
Browse by
Recent Submissions
-
Hand-designed local image descriptors vs. Off-the-shelf CNN-based features for texture classification: An experimental comparison
(2018-01-01)© Springer International Publishing AG 2018. Convolutional Neural Networks have proved extremely successful in object classification applications; however, their suitability for texture analysis largely remains to be ... -
Bayesian Torrent Classification by File Name and Size Only
(Journal of Machine Learning Research, 2016-09-09)Torrent traffic, much of which is assumed to be illegal downloads of copyrighted content, accounts for up to 35% of internet downloads. Yet, the process of classification and identification of these downloads is unclear, ... -
Moving from big data and machine learning to smart data and causal modelling: a simple example from consumer research and marketing
(2015-05-05)Provides a simple illustration of why the "pure machine learning from big data" approach is inevitably inadequate without expert judgement. Uses a causal Bayesian network -
Designing for psychological change: individuals' reward and cost valuations in weight management.
(2014-06-26)BACKGROUND: Knowledge of the psychological constructs that underlie behavior offers valuable design opportunities for persuasive systems. We use the decision theory, which describes how behavior is underpinned by reward-cost ... -
On-the-fly feature importance mining for person re-identification
(2014-04-01)State-of-the-art person re-identification methods seek robust person matching through combining various feature types. Often, these features are implicitly assigned with generic weights, which are assumed to be universally ... -
Learning Multimodal Latent Attributes
(2014-02)