192815 research outputs found
Sort by
Automated Bitcoin trading dApp using price prediction from a deep learning model
Get PDFDistributed ledger technology (DLT) and cryptocurrency have revolutionized the financial landscape and relevant applications, particularly in investment opportunities. Despite its growth, the market’s volatility and technical complexities hinder widespread adoption. This study proposes a cryptocurrency trading system powered by advanced machine learning (ML) models to address these challenges. By leveraging random forest (RF), long short-term memory (LSTM), and bi-directional LSTM (Bi-LSTM) models, the cryptocurrency trading system is equipped with strong predictive capacity and is able to optimize trading strategies for Bitcoin. The up-to-date price prediction information obtained by the machine learning model is incorporated by custom oracle contracts and is transmitted to portfolio smart contracts. The integration of smart contracts and on-chain oracles ensures transparency and security, allowing real-time verification of portfolio management. The deployed cryptocurrency trading system performs these actions automatically without human intervention, which greatly reduces barriers to entry for ordinary users and investors. The results demonstrate the feasibility of creating a cryptocurrency trading system, with the LSTM model achieving a return on investment (ROI) of 488.74% for portfolio management during the duration of 9 December 2022 to 23 May 2024. The ROI obtained by the LSTM model is higher than the performance of Bitcoin at 234.68% and that of other benchmarking models with RF and Bi-LSTM over the same timeframe. This approach offers significant cost savings, transparent portfolio management, and a trust-free platform for investors, paving the way for broader cryptocurrency adoption. Future work will focus on enhancing prediction accuracy and achieving greater decentralization
Influences of diet quality and nursery-habitat complexity and sex on brain development and cognitive performance of brown trout (Salmo trutta L.)
Get PDFAccess to omega-3 long-chain polyunsaturated fatty acids (n-3 LC-PUFA) and habitat complexity have been proposed to influence brain development and cognitive ability. We aimed to investigate the physiological and cognitive effects of dietary n-3 LC-PUFA deprivation on juvenile brown trout (Salmo trutta L.) in complex habitats resembling natal stream conditions in which populations have evolved. We tested effects of n-3 LC-PUFA deficiency in diet and habitat complexity on somatic growth, cognitive performance, encephalization, n-3 LC-PUFA biosynthesis and nutrient routing capacity. Brown trout were raised from egg for 7 months post-hatch on either a high (8.91%) or low (1.79%) n-3 LC-PUFA diet; for the final 3 months, trout were further divided into complex (heavily ornamented tanks with small, dynamic, populations) or simple habitats (bare tanks with many, constant, inhabitants). Recognition, memory and inference were tested by comparing the times required to establish stable hierarchical relationships in agonistic dyadic trials featuring naïve trout and trials in which one of the trout had previously observed the other. Gas chromatography and compound-specific stable hydrogen isotope analysis revealed increased biosynthesis and routing of n-3 LC-PUFA to the brain among trout on n-3 LC-PUFA-deficient diets. Fed to satiation, trout did not sacrifice somatic growth to fuel biosynthesis and routing of n-3 LC-PUFA. However, dietary deficiency in n-3 LC-PUFA did lead to smaller brains, and smaller brains were associated with lower cognitive performance. Complex habitats elicited better cognitive performance, and were associated with lower somatic growth, but habitat complexity played only minor roles in encephalization and the n-3 LC-PUFA composition of brain lipids. We conclude that developmental plasticity in response to environment allows brown trout partially to compensate for the paucity of dietary n-3 LC-PUFA, and we suggest that cognitive divergences may play a role in the diversification of life-history variants among brown trout in the wild
Global diversity patterns are explained by diversification rates and dispersal at ancient, not shallow, timescales
Get PDFExplaining global species richness patterns is a major goal of evolution, ecology, and biogeography. These richness patterns are often attributed to spatial variation in diversification rates (speciation minus extinction). Surprisingly, prominent studies of birds, fish, and plants have reported higher speciation and/or diversification rates at higher latitudes, where species richness is lower. We hypothesize that these surprising findings are explained by the focus of those studies on relatively recent macroevolutionary rates, within the last ~20 million years. Here, we analyze global richness patterns among 10,213 squamates (lizards and snakes) and explore their underlying causes. We find that when diversification rates were quantified at more recent timescales, we observed mismatched patterns of rates and richness, similar to previous studies in other taxa. Importantly, diversification rates estimated over longer timescales were instead positively related to geographic richness patterns. These observations may help resolve the paradoxical results of previous studies in other taxa. We found that diversification rates were largely unrelated to climate, even though climate and richness were related. Instead, higher tropical richness was related to ancient occupation of tropical regions, with colonization time the variable that explained the most variation in richness overall. We suggest that large-scale diversity patterns might be best understood by considering climate, deep-time diversification rates, and the time spent in different regions, rather than recent diversification rates alone
Investigating the behavioural responses of sheep used for teaching veterinary undergraduates
Get PDFBackground:
The use of animals for teaching veterinary medicine to veterinary students could induce stress and therefore affect the welfare of these animals.
Methods:
Six Easycare ewes were used for clinical examination classes for veterinary students once a week, for 5 consecutive weeks. Sheep behaviour was video recorded before, during and after these classes, and subsequently categorised into maintenance or stress behaviour. Data were analysed to describe behaviour observed before, during and after teaching sessions, to investigate if habituation to stress developed over the 5 weeks and to explore what factors (day, teaching session within the day, number of people in the pen) influenced stress behaviour.
Results:
Individual differences in both the number of stress behaviours exhibited and the overall stress response were observed between sheep. Overall, the total number of stress behaviours observed decreased between Day 1 and Day 5. Behaviour seen after teaching varied between days, with sheep spending more time lying on Days 2 and 3 and more time eating on Days 4 and 5. Sheep spent less time standing on all days after being used for teaching.
Limitations:
The individual differences between sheep, in combination with the low number of sheep and the short observational time, may have affected the outcomes of the study.
Conclusion:
Sheep experience stress when used in clinical examination classes, with variation between individuals. The lower number of stress behaviours observed in the last 2 days may suggest habituation to these classes. The increase in lying behaviour after teaching may indicate potential tiredness
Myopericarditis due to <i>Capnocytophaga canimorsus</i> infection after a dog bite: A case report
No full textCapnocytophaga canimorsus infection occurs after a dog/cat bite and asplenic patients are most vulnerable. We describe a rare case of myopericarditis due to C. canimorsus infection. A 41-year-old man presented with chest pain and ST-segment elevation to the coronary reperfusion service. He had been unwell for 3 days with fever and diarrhoea. Coronary angiography excluded coronary plaque rupture or occlusion, and transthoracic echocardiography revealed biventricular systolic dysfunction. Cardiac MRI confirmed myopericarditis with extensive circumferential midwall/epicardial late gadolinium enhancement (LGE) and hyposplenia. C. canimorsus was grown in blood cultures. The patient later revealed he had been bitten by a dog. Ventricular function normalised after 8 days. Myocarditis due to C. canimorsus has only been reported in an autopsy case series and one previous case report. We postulate previous cases of ‘myocardial infarction’ may have been cases of myocarditis. Cardiac MRI pattern of LGE was also helpful in risk stratification
Catchment-based approach and formalisation of artisanal and small-scale mining for sustainable mining management in the Philippines
Get PDFNo abstract available
Measurement of the ψ (2S) to J/ψ cross-section ratio as a function of centrality in PbPb collisions at √sNN = 5. 02 TeV
Get PDFThe ratio of prompt production cross-sections of ψ(2S) and J/ψ mesons in their dimuon final state is measured as a function of centrality, using data collected by the LHCb detector in PbPb collisions at √sNN = 5.02 TeV, for the first time in the forward rapidity region. The measured ratio shows no dependence on the collision centrality, and is compared to the latest theory predictions and to the recent measurements in literature
First measurement of neutron capture multiplicity in neutrino-oxygen neutral-current quasielasticlike interactions using an accelerator neutrino beam
Get PDFWe report the first measurement of neutron capture multiplicity in neutrino-oxygen neutral-current quasielasticlike interactions at the gadolinium-loaded Super-Kamiokande detector using the T2K neutrino beam, which has a peak energy of about 0.6 GeV. A total of 30 neutral-current quasielasticlike event candidates were selected from T2K data corresponding to an exposure of 1.76 ×1020 protons on target. The y ray signals resulting from neutron captures were identified using a neural network. The flux-averaged mean neutron capture multiplicity was measured to be 1.37±0.33 (stat.)+0.17
−0.27 (syst.), which is compatible within 2.3 sigma than predictions obtained using our nominal simulation. We discuss potential sources of systematic uncertainty in the prediction and demonstrate that a significant portion of this discrepancy arises from the modeling of hadron-nucleus interactions in the detector medium
SelfieAvatar: Real-time Head Avatar reenactntment from a Selfie Video
No full textHead avatar reenactment focuses on creating animatable personal avatars from monocular videos, serving as a foundational element for applications like social signal understanding, gaming, human-machine interaction, and computer vision. Recent advances in 3D Morphable Model (3DMM)-based facial reconstruction methods have achieved remarkable high-fidelity face estimation. However, on the one hand, they struggle to capture the entire head, including non-facial regions and background details in real time, which is an essential aspect for producing realistic, high-fidelity head avatars. On the other hand, recent approaches leveraging generative adversarial networks (GANs) for head avatar generation from videos can achieve high-quality reenactments but encounter limitations in reproducing fine-grained head details, such as wrinkles and hair textures. In addition, existing methods generally rely on a large amount of training data, and rarely focus on using only a simple selfie video to achieve avatar reenactment. To address these challenges, this study introduces a method for detailed head avatar reenactment using a selfie video. The approach combines 3DMMs with a StyleGAN-based generator. A detailed reconstruction model is proposed, incorporating mixed loss functions for foreground reconstruction and avatar image generation during adversarial training to recover high-frequency details. Qualitative and quantitative evaluations on self-reenactment and cross-reenactment tasks demonstrate that the proposed method achieves superior head avatar reconstruction with rich and intricate textures compared to existing approaches
Can We Edit LLMs for Long-Tail Biomedical Knowledge?
No full textKnowledge editing has emerged as an effective approach for updating large language models (LLMs) by modifying their internal knowledge. However, their application to the biomedical domain faces unique challenges due to the long-tailed distribution of biomedical knowledge, where rare and infrequent information is prevalent. In this paper, we conduct the first comprehensive study to investigate the effectiveness of knowledge editing methods for editing long-tail biomedical knowledge. Our results indicate that, while existing editing methods can enhance LLMs’ performance on long-tail biomedical knowledge, their performance on long-tail knowledge remains inferior to that on high-frequency popular knowledge, even after editing. Our further analysis reveals that long-tail biomedical knowledge contains a significant amount of one-to-many knowledge, where one subject and relation link to multiple objects. This high prevalence of one-to-many knowledge limits the effectiveness of knowledge editing in improving LLMs’ understanding of long-tail biomedical knowledge, highlighting the need for tailored strategies to bridge this performance gap
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇