--- tags: HCMUT data_mining --- Timeseries Classification === ## Refs http://www.timeseriesclassification.com/description.php?Dataset=Crop https://towardsdatascience.com/time-series-classification-using-dynamic-time-warping-61dcd9e143f6 https://towardsdatascience.com/deep-learning-for-time-series-classification-inceptiontime-245703f422db https://www.analyticsvidhya.com/blog/2019/01/introduction-time-series-classification/ https://towardsdatascience.com/a-brief-introduction-to-time-series-classification-algorithms-7b4284d31b97 ## I. Timeseries data (0.5) (Khang + Khang) ## II. Algorithms (2) (Khanh + Khanh) ### Feature base (Quang Khanh) Features can be extracted **globally** (over the entire time series) or **locally** (over regular intervals/bins, random intervals, sliding windows of intervals, and so on). Series can be transformed into **primitive values** (e.g. mean, standard deviation, slope) or into other series (e.g. Fourier transform, series of fitted auto-regressive coefficients). - https://www.arxiv-vanity.com/papers/1709.08055/#:~:text=A%20feature%2Dbased%20approach%20to,of%20time%20series%20%5B35%5D%20. - https://jwcn-eurasipjournals.springeropen.com/articles/10.1186/s13638-020-1661-4 - Interval-based Classifiers - Dictionary-Based Classification - Bag of SFA Symbols (BOSS) - The BOSS Ensemble - Contractable BOSS (cBOSS) - Frequency-based - Shapelet-Based Classifiers #### Distance base (Duy Khanh) ``` use distance metrics to determi ne class membership. https://arxiv.org/abs/1806.04509 ## DTW - https://medium.com/walmartglobaltech/time-series-similarity-using-dynamic-time-warping-explained-9d09119e48ec - https://rtavenar.github.io/blog/dtw.html - https://staff.washington.edu/dbp/PDFFILES/FCS-overheads.pdf ## DWT https://ataspinar.com/2018/12/21/a-guide-for-using-the-wavelet-transform-in-machine-learning/ https://towardsdatascience.com/the-wavelet-transform-e9cfa85d7b34 https://medium.com/financial-time-series-denoising-with-wavelet/introduction-to-wavelet-theory-ddafa4204707 https://en.wikipedia.org/wiki/Discrete_wavelet_transform http://blogs.gm.fh-koeln.de/ciop/files/2019/01/thillwavelet.pdf https://www.researchgate.net/figure/Discrete-wavelet-transform-applied-to-air-coupled-ultrasonic-probing-of-foams-From-top_fig4_245286628 ``` ![](https://i.imgur.com/zY1k9nF.png) ![](https://i.imgur.com/HfcN2g1.png) - dynamic time warping (DTW) metric: measures similarity between two sequences that may not align exactly in time, speed, or length. ### Ensemble Classifiers ### III. Applications (1.5) (Khang + Khang) - List down application (0.5) - Implement application (1)