home ⋮ blog ⋮ posts ⋮ "current trends" guest lecture

Slide 01
Slide 02
Slide 03

You can find the pbsf-lib Python module on GitHub; follow the installation instructions in the README.md.

Slide 04
Slide 05

In this introduction, we focus on univariate time series.

Slide 06

In the example above, we look at the total load on the Belgian grid over time. You can find this data at opendata.elia.be. We see some seasonality: the load on the grid is generally higher in the colder months, and drops during the warmer months.

Slide 07

Zooming in on just one month, we see a repeating pattern of higher load during workdays and lower load during weekends.

Slide 08

Overlapping all workdays of September, we see the rise in load between 3 and 6 AM, with a small dip in load around noon, and a gradual drop throughout the evening.

Slide 09
Slide 10

Slide 11
Slide 13
Slide 14
Slide 15
Slide 16

Slide 17
Slide 18
Slide 19
Slide 20
Slide 21
Slide 22
Slide 23
Slide 24
Slide 25
Slide 26

See “Time Series Data Mining Using the Matrix Profile”.

Slide 27
Slide 28
Slide 29
Slide 30
Slide 31
Slide 32
Slide 33
Slide 34
Slide 35
Slide 36

Slide 37
The more fine-grained the representation, the more accurate the distance approximation.

Slide 44

Increasing the alphabet_size solves the issue.

Slide 46
See “Sequence Learning: Paradigms, Algorithms, and Applications”.

Slide 47
Slide 48
Slide 49
Slide 50
Slide 51
Slide 55

See “Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progress” by Renjie Wu and Eamonn J. Keogh.

Slide 56
Slide 57
Slide 58
Slide 59
Slide 60
Slide 61
Slide 62
Slide 63
Slide 64
Slide 65

Note that the SlopeSignNode is just an example to explain the concept of an approximation. It generally performs poorly in approximating and comparing most time series segments. Other approximations implement a distance that lower bounds the Euclidean distance.

Slide 66
Slide 67
Slide 68

See “Hierarchical pattern matching for anomaly detection in time series” by M. Van Onsem et al.

Slide 69

See the pbsf-lib documentation on how to visualise results.

reply to this post