Models implemented for time series classification.
The demonstrator that depicts the overall workflow is present in another repo https://github.com/Adarsh-g-s/Tune-2.git
A distress measuring device has been used to capture the stress signal of a user, during the course of a task. The captured stress signal has been used to build a model which will be used to classify the difficulty level of a different task as easy or hard. A framework has been designed and implemented to realize the same. The framework comprises of two experiments - Phase I and Phase II. In Phase I participants are assigned the task of answering questions through a software and in Phase II participants are assigned the task of annotating tweets. The stress signal of the participant obtained in Phase I is used to build the classification models and they are used to classify the stress signal of the participant in Phase II in terms of easy or hard. Two categories of models have been built, non-shapelet based and shapelet based.
- SAX-VSM
- RotF
- BOSS
- SVM
- Hive COTE