Ph.D. or MSc. position for prediction of non linear time series
Publicat: 24 Sep 2015 | Vizualizari: 682
The School of Computer Science, University of Windsor, is inviting application for fully funded MSc. or Ph.D. studentships in the field of Computer Science.
The successful candidate will be working with a multidisciplinary group whose research falls within the disciplines of artificial Intelligence, chaos theory, and health informatics. The successful candidate(s) will be supervised Dr. Robin Gras. The project is funded through multiple NSERC grants.
We are looking for a talented and highly motivated individual to implement and extend the P&H method for measuring chaos level. This method is a new and efficient method for detecting the random signals from deterministic signal and it is based on the Poincar section and the Higuchi fractal dimension. This method can be used to detect chaotic behaviour in a signal.
This method recently has been applied to some biomedical, ecological and financial data. A method, GenericPred, has been developed based on the P&H measure to predict the future states of complex non-linear time series. This method provides a first step towards accurate and comprehensive time series long-term predictions. Although with respect to long-term predictions it is impossible to predict the exact values, GenericPred's performance shows great potential for predicting the time series' trend.
The successful candidate will extend the method to make the computation of the P&H measure more generic. Optimization methods will be applied and analysis of the behavior of the measure on many different conditions will be performed. New applications of the predictive methods will be developed in particular for multiple diseases diagnosis.
În acest ”Test” este vorba cât de bine poți gândi. Deoarece unii greșesc la cele mai simple întrebări.Acesta este un așa numit „Test de Logică”.
Firma: S.C. SANADOR S.R.L.
Nivel cariera: Fara experienta
Tipul postului: Full-time
Perioada de valabilitate: 2021-10-21 00:00:00 - 2021-11-06 00:00:00