Clustering for Automated Exploratory Pattern Discovery in Animal Behavioral Data

Abstract

Traditional methods of data analysis in animal behavior research are usually based on measuring behavior by manually coding a set of chosen behavioral parameters, which is naturally prone to human bias and error, and is also a tedious labor-intensive task. Machine learning techniques are increasingly applied to support researchers in this field, mostly in a supervised manner: for tracking animals, detecting land marks or recognizing actions. Unsupervised methods are increasingly used, but are under-explored in the context of behavior studies and applied contexts such as behavioral testing of dogs. This study explores the potential of unsupervised approaches such as clustering for the automated discovery of patterns in data which have potential behavioral meaning. We aim to demonstrate that such patterns can be useful at exploratory stages of data analysis before forming specific hypotheses. To this end, we propose a concrete method for grouping video trials of behavioral testing of animal individuals into clusters using a set of potentially relevant features. Using an example of protocol for testing in a "Stranger Test", we compare the discovered clusters against the C-BARQ owner-based questionnaire, which is commonly used for dog behavioral trait assessment, showing that our method separated well between dogs with higher C-BARQ scores for stranger fear, and those with lower scores. This demonstrates potential use of such clustering approach for exploration prior to hypothesis forming and testing in behavioral research.

Keywords: Data Science; animal behavior; behavioral testing; clustering; machine learning.

Figure 1

A high level overview of the general approach.

Figure 2

(A) detection; (B) trajectory extraction.

Figure 3

Descriptive statistics (before normalization and scaling).

Figure 4

Patterns of Scenario 1: Cluster 0 red dots, Cluster 3 green dots.

Figure 5

Results of clustering scenario 1 along the axes of total contact and duration of approach.

Figure 6

Results of clustering scenario 2 along the axes of contact ratio and area.

Figure 7

C-BARQ factors descriptive statistics for scenario 1.

Figure 8

SDF comparison between C0 and C3 (scenario 1), dotted line is the median, solid line in the box is the mean.

Figure 9

SDA comparison between C0 and C3 (scenario 1), dotted line is the median, solid line in the box is the mean.

Figure 10

PS comparison between C0 and C3 (scenario 1), dotted line is the median, solid line in the box is the mean.