Dataset Information

BARN: Behavior-Aware Relation Network for multi-label behavior detection in socially housed macaques.

ABSTRACT: Quantification of behaviors in macaques provides crucial support for various scientific disciplines, including pharmacology, neuroscience, and ethology. Despite recent advancements in the analysis of macaque behavior, research on multi-label behavior detection in socially housed macaques, including consideration of interactions among them, remains scarce. Given the lack of relevant approaches and datasets, we developed the Behavior-Aware Relation Network (BARN) for multi-label behavior detection of socially housed macaques. Our approach models the relationship of behavioral similarity between macaques, guided by a behavior-aware module and novel behavior classifier, which is suitable for multi-label classification. We also constructed a behavior dataset of rhesus macaques using ordinary RGB cameras mounted outside their cages. The dataset included 65 913 labels for 19 behaviors and 60 367 proposals, including identities and locations of the macaques. Experimental results showed that BARN significantly improved the baseline SlowFast network and outperformed existing relation networks. In conclusion, we successfully achieved multi-label behavior detection of socially housed macaques with both economic efficiency and high accuracy.

SUBMITTER: Yang S

PROVIDER: S-EPMC10802107 | biostudies-literature | 2023 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

BARN: Behavior-Aware Relation Network for multi-label behavior detection in socially housed macaques.

Yang Sen S Chen Zhi-Yuan ZY Liang Ke-Wei KW Qin Cai-Jie CJ Yang Yang Y Fan Wen-Xuan WX Jie Chen-Lu CL Ma Xi-Bo XB

Zoological research 20231101 6

Quantification of behaviors in macaques provides crucial support for various scientific disciplines, including pharmacology, neuroscience, and ethology. Despite recent advancements in the analysis of macaque behavior, research on multi-label behavior detection in socially housed macaques, including consideration of interactions among them, remains scarce. Given the lack of relevant approaches and datasets, we developed the Behavior-Aware Relation Network (BARN) for multi-label behavior detection ...[more]

PMID: 37804114

Similar Datasets

Project description:BackgroundPeaceful conflict resolution strategies have been identified as effective mechanisms for minimising the potential costs of group life in many gregarious species, especially in primates. The knowledge of conflict-management in orangutans, though, is still extremely limited. Given their semi-solitary lives in the wild, there seems to be barely a need for orangutans to apply conflict management strategies other than avoidance. However, because of the rapid loss of orangutan habitat due to deforestation, opportunities to prevent conflicts by dispersion are shrinking. Additionally, more and more orangutans are brought into rehabilitation centres where they are bound to live in close contact with conspecifics. This raises the questions of whether and how orangutans are able to cope with conflicts, which are inevitably connected with group life.MethodsObservational zoo-studies provide a valuable method to investigate such potential: in zoos, orangutans usually live in permanent groups and face the challenges of group life every day. Therefore, we observed a group of six socially-housed Sumatran orangutans at the Dortmund Zoo, Germany, both in their spacious outdoor enclosure in the summer and in the less spacious indoor enclosure in the winter. During 157.5 h of observation, we collected data on aggressive interactions, third-party interventions and post-conflict affiliations. We applied the post-conflict/matched-control observation (PC/MC) and the time rule method to investigate the occurrence of reconciliation and post-conflict third-party affiliations.ResultsWe recorded a total of 114 aggressive interactions (including conflicts in the context of weaning and of male sexual coercion). As expected, we found an increase of both open conflicts and peaceful conflict resolution under less spacious conditions. In accordance with previous reports, we observed interventions by initially uninvolved individuals. Whereas we found no clear evidence for post-conflict third-party affiliations, we were able to demonstrate the occurrence of reconciliation among orangutans.DiscussionNotwithstanding the small sample size and the explorative character of our study, we found evidence that orangutans possess a potential for prosocial conflict resolution. When living in groups and under conditions in which dispersion is no longer an option, orangutans are capable to flexibly apply strategies of conflict resolution to cease open conflicts and to repair the potential social damage of aggressive interactions. These strategies are similar to those of other great apes.

Project description:The compost-bedded pack barn (CBPB) system has been increasingly adopted by dairy farms due to its ability to enhance animal comfort and milk production. This study evaluated the associations among bedding characteristics, milk quality and composition, and subclinical mastitis (SCM) occurrence in dairy herds housed in CBPB systems. Over a period of six months, data related to milk quality and udder health and bedding sampling were collected from eight dairy farms. Monthly measurements of the bedding temperature and wind speed inside the CBPB were taken, while temperature and relative humidity data inside the CBPB were recorded using a datalogger. Bedding samples were subjected to analysis of moisture, pH, microbiological count, and carbon/nitrogen ratio. Data on milk composition (fat, protein, milk urea nitrogen, and total solids) and quality (somatic cell count and standard plate count) of bulk tank milk were obtained from DHIA results. Canonical correlation analyses were used to evaluate the association between the analyzed group variables, and linear regression models were used to identify associations between bedding characteristics and SCM occurrence in the studied herds. The bedding characteristics that most influenced milk composition and quality were moisture, temperature at 30 cm depth (T30), and bedding pH. Environmental variables played an important role in bedding composting, as they were closely related to the surface temperature and pH. Overall, 62.71% of the variation in milk quality and composition could be explained by the bedding variables, and 77.50% of the variation in the bedding variables was associated with environmental variables. Median SCM prevalence and incidence were 28.6 and 13.8%, respectively. An increase of 1 °C for T30 resulted in a 0.6% reduction in the prevalence of SCM. Additionally, the bedding surface temperature at 22.3 °C resulted in the highest incidence of SCM (~18.1%). Our results demonstrate the importance of controlling microclimatic conditions in the CBPB to optimize the bedding composting process and milk quality.

Project description:Unhealthy eating behavior is a major public health issue with serious repercussions on an individual's health. One potential solution to overcome this problem, and help people change their eating behavior, is to develop conversational systems able to recommend healthy recipes. One challenge for such systems is to deliver personalized recommendations matching users' needs and preferences. Beyond the intrinsic quality of the recommendation itself, various factors might also influence users' perception of a recommendation. In this paper, we present Cora, a conversational system that recommends recipes aligned with its users' eating habits and current preferences. Users can interact with Cora in two different ways. They can select pre-defined answers by clicking on buttons to talk to Cora or write text in natural language. Additionally, Cora can engage users through a social dialogue, or go straight to the point. Cora is also able to propose different alternatives and to justify its recipes recommendation by explaining the trade-off between them. We conduct two experiments. In the first one, we evaluate the impact of Cora's conversational skills and users' interaction mode on users' perception and intention to cook the recommended recipes. Our results show that a conversational recommendation system that engages its users through a rapport-building dialogue improves users' perception of the interaction as well as their perception of the system. In the second evaluation, we evaluate the influence of Cora's explanations and recommendation comparisons on users' perception. Our results show that explanations positively influence users' perception of a recommender system. However, comparing healthy recipes with a decoy is a double-edged sword. Although such comparison is perceived as significantly more useful compared to one single healthy recommendation, explaining the difference between the decoy and the healthy recipe would actually make people less likely to use the system.

Dataset Information

BARN: Behavior-Aware Relation Network for multi-label behavior detection in socially housed macaques.

Publications

BARN: Behavior-Aware Relation Network for multi-label behavior detection in socially housed macaques.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets