Seminar by Ms Monika Berendsen, "Software Development at EVONIK"

Seminar by Ms Monika Berendsen, Evonik Industries AG, entitled "Software Development at EVONIK"

 

Location: P302, Poole House, Bournemouth University.

Seminar by Mr Alexander Maslov, "Quantile Index for Gradual and Abrupt Change Detection in Noisy Sensor Data Monitoring"

Seminar by Mr Alexander Maslov, Information Systems Group, Eindhoven University of Technology, Holland, entitled "Quantile Index for Gradual and Abrupt Change Detection in Noisy Sensor Data Monitoring"

Seminar by Dr Georg Krempl, "Online Clustering of High-Dimensional Trajectories under Concept Drift"

Seminar by Mr Dr Georg Krempl, Knowledge Management and Discovery group at the University of Magdeburg, Germany, entitled "Online Clustering of High-Dimensional Trajectories under Concept Drift"

ECMLPKDD'12 Tutorial: Advanced Topics in Data Stream Mining, Bristol, UK

Event Date: 
24 Sep 2012 (All day)

Advanced Topics in Data Stream Mining

ECMLPKDD'12 Tutorial

Bristol UK, 24 September 2012

The tutorial will provide an overview of the state-of-the-art advances in mining streaming data. The first part introduces stream mining under concept drift and covers methods and tools for classification, regression and frequent pattern mining. The second part of the tutorial addresses the question of mining multiple streams that are independent or interdependent. Concept drift plays a central role in this tutorial: we address it in the context of conventional one-stream mining to set the scene, and recapitulate on it after introducing multiple-stream mining. For concept drift, we also consider machine learning methods that are appropriate for incremental data and slow streams.

Tutorialists

Seminar by Mr. Aleksander Badura & Mr Tobiasz Dworak: "Introduction to System Testing”. Wednesday the 22nd of August, 2012.

In this talk, the speakers will provide an introduction to Software Testing with focusing on Test level or System Test. They will discuss how testing within a project Life Cycle is planned and performed; and the seminar will conclude I with an example of testing within INFER project.

Our gests Mr. Aleksander Badura & Mr Tobiasz Dworak, have extensive experience in Software System's Architecture Design, Development and Testing in a wide range of projects within the international company Research & Engineering Centre (REC-global).
 

One-day symposium on Signal Processing and Kernel Methods (next Tuesday the 26th at 15h 30’ in PG16 Theatre)

A short one-day symposium on Signal Processing and Kernel Methods will take place next Tuesday, the 26th of June in PG 16 Lecture Theatre at 15:30 h (Ground floor, Poole House), organized in the context of the INFER project.

The symposium will have the participation of academics of the Smart Technology Research Center in Bournemouth University and  of three members of the Image and Signal Processing Group of University of Valencia, Spain. This world-leading group is particularly influential in several Machine Learning  areas and applications like for instance Image Processing (in Geosciences, Medical Brain Imaging, etc.) and Kernel Machines.


The agenda of the event will be the following:
•    15h 30’. Short intro by Dr. Malo:  "Research at the Image and Signal Processing Group". Jesús Malo. A brief overview of our research interests and lines.
•    15h 40’. Dr Laparra: "Gaussianization Framework for Signal Processing”
Abstract: We generalize a class of projection pursuit methods to transform arbitrary multidimensional data into multivariate normal data, thus attaining statistical independence of its components. The proposed analysis enables a number of novel ways to solve practical problems in high-dimensional scenarios, such as those encountered in image processing, speech recognition, array processing, or bioinformatics. Our framework extends Independent and Principal Components Analyses-based methods, which are typically not applicable to data generated from nonlinear, non-independent or non-Gaussian sources. The performance is successfully illustrated in a number of multidimensional data processing problems such as image synthesis, classification, saliency analysis, and de-noising.
•    16h.Dr. Camps-Valls:  "Extended Kernel Methods".
Abstract: I will talk about our love story with kernel methods for the last 10 years. Kernel methods constitute a simple way of translating linear algorithms into nonlinear ones. I will revise several interesting developments for 1) time series analysis, regression and function approximation; 2) classification problems; 3) nonlinear feature extraction; and 4) dependence estimation. The introduced methods extend previous standard algorithms to deal with non-stationary environments and structured domains, and the presence of non-Gaussian noise. Additionally, I'll briefly talk about a way to learn the kernel function directly from the data via clustering or graphs. Examples in signal and image processing will guide this overview.
•    16h 30’. Discussion Panel by Dr. Camps, Dr. Malo,, Dr Laparra, Prof. Gabrys, Dr. Bouchachia, Dr. Balaguer-Ballester and any researchers who would like to join the discussion.
 

Live broadcasting of Seminar: "Current state of INFER platform software (Hands on Demo)

Talk broadcasted live via Skype to all Centers involved on INFER.  Mr Tobiasz Dworak will talk about Software Platforms for Evolving Predictive Systems. The talk will take place on Wednesday, 13th of June in Lawrence Lecture Theater at 14:00 h.

Seminar: Introduction to Adaptive Learning from Streaming Data by Dr Zliobaite

Our next speaker of the STRC seminar series will be Dr Indrė Žliobaitė.The talk will take place next Wednesday, 30th of May in Lawrence Lecture Theatre at 16:00 h (please click for a map)

Indrė (Lecturer in BU as most of us know) will present novel angles of her work in a highly didactic fashion. She will talk about an exciting topic, strategies for predicting streaming data. This is particularly attractive for instance for those of us involved in projects in real-time industrial settings.

Please feel free to show up if you like it regardless you background!

 The talk title is:

“Introduction to Adaptive Learning from Streaming Data”

Short description:

Changing data over time presents one of the major challenges in predictive modelling applications, for example automated movie recommendation, bankruptcy prediction, spam categorization, food sales prediction and many more. In such situations predictive models need to have mechanisms to update or retrain themselves using recent data, otherwise they will quickly lose accuracy. This talk will give an introductory overview of settings and algorithms for adaptive predictive modelling.

Posters accepted in the 4th Annual PGR Conference by Rashid Bakirov et al. & Manuel Martin-Salvador et al.

Two poster presentations by members of the INFER team have been accepted in the 4th Annual Post-Graduate Researchers Conference at Bournemouth University on the 28th of June, 2012.
The titles are:
•    “Automatic and Adaptive Preprocessing for the Development of Predictive Models" by Mr Manuel Martin-Salvador, directed by Dr. Indre Zilobaite & Porf. Bodgan Gabrys.
•     “Multielement Adaptive Models for Learning with Streaming Data: Motivation and Vision” by Mr. Rashid Bakirov, directed by Dr. Indre Zilobaite & Porf. Bodgan Gabrys.
Congratulations to the authors!
 

New accepted paper: Next challenges for adaptive learning systems

New paper which emerged from a workshop in Adaptive Learning Systems in the context of INFER. Soon will appear in the publications list. Warm congratulations to the authors!

Syndicate content