A Significance Test for Inferring Affiliation Networks from Spatio-Temporal Data.
Bottom Line:
In this article, we provide a statistical approach to the problem of inferring the social structure of a group from the movement patterns of its members.By constructing an appropriate model, we are able to construct a significance test to detect meaningful affiliations between members of the group.We demonstrate our method on large-scale real-world data sets of positional data of flocks of Merino sheep, Ovis aries.
View Article:
PubMed Central - PubMed
Affiliation: Department of Computer Science, University College London, London, United Kingdom.
ABSTRACT
Scientists have long been interested in studying social structures within groups of gregarious animals. However, obtaining evidence about interactions between members of a group is difficult. Recent technologies, such as Global Positioning System technology, have made it possible to obtain a vast wealth of animal movement data, but inferring the underlying (latent) social structure of the group from such data remains an important open problem. While intuitively appealing measures of social interaction exist in the literature, they typically lack formal statistical grounding. In this article, we provide a statistical approach to the problem of inferring the social structure of a group from the movement patterns of its members. By constructing an appropriate model, we are able to construct a significance test to detect meaningful affiliations between members of the group. We demonstrate our method on large-scale real-world data sets of positional data of flocks of Merino sheep, Ovis aries. No MeSH data available. |
Related In:
Results -
Collection
License getmorefigures.php?uid=PMC4508121&req=5
Mentions: To check the performance of the significance test we calculate the connectivity levels both within and between the two sub-flocks. (The term connectivity level refers to the proportion of edges present in comparison to the total number of possible edges. For example, a connectivity level of one indicates that all possible edges are present.) The results are given in Fig 7. As expected, the level of connectivity within the two sub-flocks is far higher than between the two sub-flocks. The proportion of connections between the two flocks (i.e. the false-positive rate) was 4.9 ± 1.6%, which is slightly higher than expected. An explanation for this is that, even though we expect the behaviour of the two sub-flocks to be disparate, certain heterogeneous aspects of the data are causing a higher false-positive rate than expected. An illustration of this fact is given in Fig 8. It can be seen that the animals spend a disproportionately large amount of time around the entrance of the field, which is where the drinking trough was located. Given this heterogeneous aspect of the data, we applied the significance test to an irregular partition of the sample space, with all other aspects of the test remaining the same. In particular, we used algorithm 1 in Table 1 to construct a partition of the sample space such that roughly the same amount of observations fell into each of the subregions of the partition. The results from using this partition are given in Fig 7, with a false-positive rate of 2.6 ± 0.6%. This is significantly lower than the false-positive rate obtained when using a regular partition. This result shows the benefit of using irregular partitions of this form when the data sets is highly heterogeneous. |
View Article: PubMed Central - PubMed
Affiliation: Department of Computer Science, University College London, London, United Kingdom.
No MeSH data available.