Multivariate time series classification codes and data sets are onlineSaturday, 03 November 2012 13:00 The codes and data sets for our paper Multivariate Time Series Classification with Learned Discretization are online. Please find the details by clicking the link. SMTS (Symbolic Multivariate Time Series) discretizes the observation space in a supervised manner to obtain the symbolic representation for classification. It is mostly implemented in R (uses the randomForest package) and C (time consuming for loops are in C). There is no explicit feature extraction, the features are learned into symbolic representation. It can handle nominal (categorical) time series and missing values. It is multiclass (does not require training multiple models as in Support Vector Machines (SVM)). It scales well with number of features (variables) and the number of time series. A treebased ensemble (Random forest) is used to learn the symbols. Two parameters are important: Alphabet size and number of trees to generate the symbolic representation. Since each tree is trained on random subsample of the instances and features, different views of the same time series are represented by the ensemble (has some connection to scalespace theory). The codes of SMTS are available on http://www.mustafabaydogan.com/files/viewcategory/14multivariatetimeseriesclassification.html. Please let me know if you have any questions by contacting me through the contact link in the menu above. {jcomments on}
The presentation of TSPD is uploadedWednesday, 17 October 2012 14:16 The presentation of TSPD in the INFORMS'12 conference in Phoenix is uploaded. You can find it on http://www.mustafabaydogan.com/files/viewcategory/8presentations.html. Please let me know if you have any questions! {jcomments on} TSPD codes are now availableMonday, 08 October 2012 10:50 The second study of my dissertation, Supervised Time Series Pattern Discovery through Local Importance (TSPD), has been submitted to Knowledge and Information Systems. The codes and details are available on Supervised Time Series Pattern Discovery through Local Importance (TSPD). Please contact me if you have any questions. {jcomments on} Our results are onlineTuesday, 02 October 2012 22:33 A webpage summarizing the error rates of the time series classifiers proposed in my dissertation is created. For now, only TSBF results (as well as the competitors') are reported. TSPD and SMTS results are coming soon! Click here. {jcomments on} 



More Articles... 

«StartPrev123NextEnd» 

Page 3 of 3 
 ► 2016 (2)
 ► 2015 (2)
 ► 2014 (6)
 ► 2013 (9)
 ► 2012 (8)