Distributed Information Systems Laboratory LSIR

Deep Learning based Pattern/Trend Mining over Time Series

Project Details

Deep Learning based Pattern/Trend Mining over Time Series

Laboratory : LSIR Semester / Master Proposal




Description:

Description

Deep learning technique is becoming a very hot topic for various machine learning and data mining problems, but utilizing it in temporal data mining, i.e. time series data mining still receives little attention.

In this project, we will try to figure out a clear road-map for exploiting recurrent neural network to discover the hidden correlation/causality among massive time series. This is a very fundamental and important task in a variety of applications. For instance, image a scenario where you observe a sudden change or pattern of interest in stock time series, a natural question is whether and to what extent some other stocks would be affected by such changes or patterns in the near future.


  • good programming skills
  • data mining/machine learning experience is a plus.

Contacts

In case of any questions, please drop us an email or come to our offices:

Site:
   
Contact: Tian Guo
Useful Links

Project Guidelines