Overview and Introduction

Topic models such as Latent Dirichlet Allocation (LDA) have been used successfully in many domains for data mining. Originally designed for text documents, these methods find some hidden “topics” considering that each document is a weighted mixture of topics. Each topic expresses itself in a document by generating some specific words with more probability than others.

Topic models have been used with various kinds of data ranging including text, image, video and mixtures of these. When applied to temporal data like videos, topic models needs to be extended to take into account the time dimension. Different approaches have been proposed towards this inclusion of the time information.

This web site proposes different methods for mining temporal data for recurrent patterns. These methods are:

• un-supervised: there is no need to annotate data,
• generic: they can be applied to various kind of data,
• temporal: the time information is capture within the models,
• probabilistic: they allow for easy interpretation and extension,
• proven: they have been used for mining various data such as video and audio,

On this website, you might be interested by:

• an introduction to temporal data mining
• different models:
  • PLSM: a temporal model that finds recurrent temporal motifs
  • HDLSM: a Bayesian non-parametric version of PLSM
  • TAMM/VLTAMM and MER: see articles on the related publications page
• the list of down-loadable content
• a list of related publications

Further models, material and code might be provided in the future.

Contact

If you want further information and comments about this website, please contact the followin persons: