Networks of trophic links (food webs) are used to describe and understand mechanistic routes for translocation of energy (biomass) between species. However, a relatively low proportion of ecosystems have been studied using food web approaches due to difficulties in making observations on large numbers of species. In this paper we demonstrate that Machine Learning of food webs, using a logic-based approach called A/ILP, can generate plausible and testable food webs from field sample data. Our example data come from a national-scale Vortis suction sampling of invertebrates from arable fields in Great Britain. We found that 45 invertebrate species or taxa, representing approximately 25% of the sample and about 74% of the invertebrate individuals included in the learning, were hypothesized to be linked. As might be expected, detritivore Collembola were consistently the most important prey. Generalist and omnivorous carabid beetles were hypothesized to be the dominant predators of the system. We were, however, surprised by the importance of carabid larvae suggested by the machine learning as predators of a wide variety of prey. High probability links were hypothesized for widespread, potentially destabilizing, intra-guild predation; predictions that could be experimentally tested. Many of the high probability links in the model have already been observed or suggested for this system, supporting our contention that A/ILP learning can produce plausible food webs from sample data, independent of our preconceptions about “who eats whom.” Well-characterised links in the literature correspond with links ascribed with high probability through A/ILP. We believe that this very general Machine Learning approach has great power and could be used to extend and test our current theories of agricultural ecosystem dynamics and function. In particular, we believe it could be used to support the development of a wider theory of ecosystem responses to environmental change.

This study aimed to use retirement data from working guide dogs to investigate healthy ageing in dogs and the demographic factors that influence ageing. Using a dataset of 7686 dogs spanning 20 years, dogs withdrawn for health reasons before they reached retirement were identified. Cases of retirement for old age, rather than for health reasons, were also recorded, as was the length of working life for all dogs. Specific health reasons were grouped into 14 different health categories. The influence of purebred or crossbreed, breed, and sex on the incidence of these health categories and the length of working life within each health category was considered. The majority (n = 6465/7686; 84%) of working guide dogs were able to function as guide dogs until they had worked for 8.5 years, when they retired. This working life might constitute a reference for the different breeds considered, with the exception of the German shepherd dog, which had a shorter working life. The most common reason for health withdrawals was musculoskeletal conditions (n = 387/1362; 28%), mostly arthritis. Skin conditions (mostly comprised of cases of atopic dermatitis) reduced working life most commonly (mean, approximately 5 years). Nervous sensory conditions (35% of which were cases of epilepsy) reduced working life by 3 years.

No abstract

The domestic dog is one of our most popular companions and longest relationships, occupying different roles, from pet to working guide dog for the blind. As dogs age different behavioural issues occur and in some cases dogs may be relinquished or removed from their working service. Here we analyse a dataset on working guide dogs that were removed from their service between 1994 and 2013. We use the withdrawal reasons as a proxy for the manifestation of undesirable behaviour. More than 7,500 dogs were in the dataset used, 83% of which were retired (due to old age) and 17% were withdrawn for behavioural issues. We found that the main reasons for behaviour withdrawal were environmental anxiety, training, and fear/aggression. Breed and sex had an effect on the odds of dogs being withdrawn under the different reasons. The age at withdrawal for the different withdrawal reasons suggested that dogs were more likely to develop fear/aggression related issues early on, whilst issues related to training could develop at almost any age. We found no evidence for heterosis effecting behaviour. We believe that this work is relevant to the pet dog population and had implications for understanding ageing and genetic influences on behaviour.