time series

Terms from Artificial Intelligence: humans at the heart of algorithms

Page numbers are for draft copy at present; they will be replaced with correct numbers when final book is formatted. Chapter numbers are correct and will not change now.

A time series is a data set wher successive data items were refer to the same phenomenon at different potns in time. These may be sampled uniformly, for example, measuring the air temperature every morning at 6am, or spordically, for example blood-pressure reading when a patient happens to have been with the doctor for various other reasons. The data may also be homgenuous, each item of identicsl form, or heterogeneous , for example wher a heakth record mayinclude all sorts of different tests at different times. There are various statisticla and AI methods for time series analysis although, in general, periodic homogeneous data is dealt with most well.

Used on Chap. 8: pages 170, 171; Chap. 12: pages 276, 277; Chap. 14: pages 319, 320, 322, 332, 333, 334, 338, 341