5 1 15 2 months 25 3 35 featurizing time series

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: enta-on of each pa-ent •  Features will serve as covariates in a regression model •  Most extracted features will be irrelevant •  Rely on model selec-on / methods robust to irrelevant features   Time Series Data •  Repeated measurements of variables over -me   ALSFRS ques-on scores   Alterna-ve ALS measures (forced and slow vital capacity)   Vital signs (weight, height, blood pressure, respiratory rate)   Lab tests (blood chemistry, hematology, urinalysis) •  Number and frequency of measurements vary across pa-ents Featuriza-on   Goal: Compact numeric representa-on of each pa-ent •  Features will serve as covariates in a regression model •  Most extracted features will be irrelevant •  Rely on model selec-on / methods robust to irrelevant features   Time Series Data •  Compute summary sta-s-cs from each -me series   Mean value, standard devia-on, slope, last recorded value, maximum value… •  Compute pairwise slopes (difference quo-ents between adjacent measurements)   Induces a deriva-ve -me series   Extract same summary sta-s-cs Featurizing Time Series Data ALSFRS Score 40 39 38 37 36 0 0.5 1 1.5 2 Months 2.5 3 3.5 Featurizing Time Series Data ALSFRS Score 40 Features extracted •  Mean = 38.75 •  SD = 0.816 •  Max = 40 •  Min = 37 •  Last = 37 •  etc. 39 38 37 36 0 0.5 1 1.5 2 Months 2.5 3 3.5 Featurizing Time Series Data ALSFRS Score 40 Features extracted •  Mean = 38.75 •  SD = 0.816 •  Max = 40 •  Min = 37 •  Last = 37 •  Slope =  ­1 •  etc. 39...
View Full Document

This note was uploaded on 02/03/2014 for the course STATS 202 taught by Professor Taylor during the Fall '09 term at Stanford.

Ask a homework question - tutors are online