30793 research outputs found
Sort by
A Risk Prediction Model for Real Estate Corporations Using High-Target Semantic BERT and Improved GRU
Get PDFAccurately predicting real estate enterprise risk is crucial for the national economy. Although some initial works have been made on this topic such as Z-score, support vector machines, and logistic regression, there remains a gap in comprehensive models that can effectively capture the dynamic risk fluctuations from real estate-specific data. As such, a novel prediction model called HRAGRU is proposed for real estate enterprises to forecast potential risk through multimodal data including news reports, policy updates, and stock information in this paper. We first extract the semantic information from news text by using a BERT model optimized for high-target semantic density. Then we investigate the relationships among various data types through a graph neural network (GNN) model with randomly masked edges or nodes. Finally, we establish an improved gated recurrent unit (GRU) model to capture the interactions between new and historical data. The effectiveness of the proposed HRAGRU model is validated using data from A-share and Hong Kong-listed real estate companies, demonstrating its superior performance in forecasting corporate risk indices. Our sources are released at https://github.com/maxiaoyan290/HRAGRU10.13039/501100012166-National Key Research and Development Program of China;
10.13039/100000001-National Science Foundation
Machine Learning of Solar Energy Forecasting Using Ensemble LSTM Method
Get PDFThe transition in energy systems aims for efficiency improvements at a higher level, targeting the reduction of climate change impacts. Investing in solar energy, endorsed by the global scientific community, is essential. One of the core obstacles hindering the seamless integration of photovoltaic (PV) systems into power grids is the associated uncertainty. This paper focuses on machine learning forecasting algorithms for solar power generation, specifically using the Long Short-Term Memory (LSTM) algorithm, a type of recurrent neural network (RNN). Accurate prediction models are crucial for maximising the efficiency and reliability of solar energy systems, especially with high PV penetration. The LSTM architecture's ability to capture temporal dependencies makes it well-suited for time series forecasting tasks such as solar irradiance prediction. Using ensemble LSTM improves output accuracy and reduces mean square error in solar energy plant production. Temporal variations in solar power production, analysed using SARIMA/ ARIMA models, highlight the volatility of PV power generation, which causes issues such as frequency instability, dispatch difficulties, and surges in current/voltage on the grid
Exploring the Effectiveness of Generative AI-Powered Chatbots in Strengthening Consumer Trust and Enhancing Retention Capabilities for a Bank
Get PDF[Report summary]The study explores the effectiveness of AI-powered chatbots in enhancing consumer trust and retention in a bank. It highlights how Generative AI based chatbots are revolutionizing customer interactions by offering personalized, efficient, and secure services. The paper discusses the integration of Generative AI in various functions, including customer service, fraud detection, and financial advice, emphasizing the role of chatbots in improving operational efficiency and customer satisfaction
Transitions to adulthood in men and women in rural Malawi in the 21st century using sequence analysis: Some evidence of delay
Get PDFBackground: Many sub-Saharan African countries have large populations of young people, and these cohorts have the potential to bring significant change. Understanding the changing lives of young people is important for ensuring individuals experience healthy and successful transitions to adulthood and for understanding how best to ensure they realise their full potential, for themselves and their communities.
Methods: This analysis used detailed, prospectively collected data on men and women from the Karonga Health and Demographic Surveillance Site in rural Malawi between 2004–2017. Multi-channel sequence analysis was conducted for the domains of leaving school, getting married, and having children. Sequences were grouped into clusters and descriptive and multinomial logistic regression carried out to assess differences in the transition to adulthood by sex, calendar time, and sociodemographic factors.
Conclusions: Many participants followed a fairly early and rapid transition to adulthood; however, a sizable proportion experienced later marriage. These participants were split into those (mostly men) who received little education and had a long delay between school-leaving and marrying, and those who spent longer in education. Access to secondary school increased over time for both men and women, while only women experienced delays in age at marriage. Where associations between sociodemographic factors and transitions to adulthood were found, they suggested that greater access to resources was associated with a later transition out of education, and therefore to marriage.
Contribution: Sequence analysis applied to longitudinal HDSS data proved a useful tool to identify different trajectories to adulthood in this rural Malawian context
Rolling Horizon-Based Operation Optimisation of Energy Storage Systems and Electric Vehicles Integrated in Distribution Networks
Get PDFEffective management algorithms have emerged to mitigate the adverse effects of high renewable energy systems and electric vehicle (EV) penetration on distribution networks. This study proposes a rolling horizon optimization-based approach to reduce power loss in distribution networks that incorporate EVs, battery-based energy storage systems (BESS) and photovoltaic (PV) plants. This approach entails an efficient charging algorithm for EVs and BESS, formulated as a mixed-integer linear programming (MILP) model and the power loss equation is linearized using a piecewise approach with the Special Ordered Sets of Type 2 (SOS2) method. The algorithm accounts for uncertainties related to EV charging and PV generation, yielding promising results. Our findings indicate that utilizing BESS and effectively managing EV charging profiles can be leveraged to reduce the power loss on the PV-integrated distribution systems. Also, the effects of the location of BESS and EV parking lot on the network are analysed thoroughly and even a decrease of 7.6% in power loss is evidenced due to the relocation of the EV parking lot between feeders.The work of Hilal Ozdemir was supported by the EPSRC Grant Reference EP/W524542/1. The work of Ozan Erdinç was supported by The Scientific and Technological Research Council of Türkiye (TUBITAK) 100th Year Science Encouragement Award
Analysis of backhaul networks
Get PDFThis thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University LondonIn its most basic form, the target of fifth generation (5G) is to provide reliable and continued connectivity for the user despite obstacles and challenges. These obstacles and challenges vary depending on the scenario, whether it’s an urban or rural area in developed or developing countries.
This thesis focuses on the 5G backhaul for standalone (SA) network and the im-pacts of backhaul technologies on the Quality of Service (QoS) and user experiences, in particular, end to end delay (E2E) and capacity planning requirements. In par-ticular, the aim is to facilitate the work of providers, developers, and investors when planning to introduce 5G technology to developing countries. This work looks into employing simulation-based approach to consider bandwidth aspects when design-ing/ upgrading current/ future cellular systems in developing countries. It presents a scheme to maximize the use of bandwidth considering both capacity and delay aspects and helps to identify major parameters that influence system design for different 5G use cases and scenarios.
The result proves that the method to determine the required link capacity is by observing the traffic delay and users access statistics as well as by increasing the capacity incrementally by changing the factor for each link in the network, un-til optimal capacity is achieved. It also indicates that within the ”broadband in the crowd” scenario for 5G services and applications, the necessary bandwidth for last-mile network connections can vary depending on the service type. Specifically, bandwidth requirements can be lessened for ultra-low latency services and applica-tions, with even greater reductions possible for those that do not require such low latency. These adjustments are observed when the backbone link is operating at its full capacity.
The finding shows that for developing countries, and by considering the cost of the bit per second, user down link/ uplink, and convenience of user terminals as more critical considerations for the adoption of one of the technologies for backhauling 5G traffic, a satellite and hybrid topology based on the existing networks, financial considerations will play an important role in determining the backhaul network topology with optimizing for the specific requirements
Considerations when applying equity weights within economic evaluation when making drug reimbursement decisions
Get PDFData availability:
No datasets were generated or analysed during the current study.When decision-makers use economic evaluation to facilitate making decisions about reimbursing whether to reimburse pharmaceuticals within a publicly funded health care system, they may consider whether to prioritise specific patient populations or diseases: e.g., cancer or rare disease. This can be achieved through applying equity weights to outcomes such as QALYs. Decision makers, however, must choose whether equity weights are applied to solely the treatment of a specific disease or to treatments of the patient with the specific disease. Without such clarification, confusion may arise which can hinder the work of those who must make reimbursement recommendations and decisions. This study examines the repercussions of implementation of equity weights. For illustration, two hypothetical case studies relating to a rare disease are considered. The first case study demonstrates that applying equity weights only to the treatment of the rare disease of interest can lead to a patient with that rare disease accruing less benefits at a higher cost to the payer. The second case study demonstrates that if equity weights are applied to the patients who have a specific rare disease, then funding of a treatment for a common disease may be restricted only to those patients for whom treatment is more costly and less effective. As discussions continue with respect to applying equity weights within economic evaluation, it is important that the repercussions outlined are recognised.No financial support for this study was provided
Effect of Tin Mining Waste as Sand Replacement on Flowability and Compressive Strength of Mortar
Get PDFEnvironmental degradation due to excessive sand mining activity and tin tailing waste disposal from mining need to be resolved. Utilization of tin tailing as sand replacement in mortar production would cut amount of waste thrown and decrease river sand consumption. The current study examines the effect of tin tailing content sand replacement on flowability and strength of mortar. Five forms of mixes were produced using different percentage of tin tailing as fine aggregate replacement. Two types of tests were conducted namely flowability test and compressive strength test. The investigation discovers that blending tin tailing affects the flowability of mortar. Utilization of up to 30% tin tailing able to produce mortar with the targeted strength. Usage of high amount of tin tailing causes declination of mortar strength. Recycling tin tailing for use in mortar production saves the use of natural sand and reduce waste disposal.We extend our appreciativeness to Universiti Malaysia Pahang Al-Sultan Abdullah for funding provided
via RDU223006. The first author also extends her thankfulness to Ministry of Higher Education for
sponsoring her study through Hadiah Latihan Persekutuan
Religion and COVID-19: methodists and Church of England followers more likely to have been vaccinated than Muslims and Pentecostals
Get PDFThis article is republished from The Conversation under a Creative Commons license. Image rights: EPA/Neil Hall; Alamy/Geoff Pugh/Daily Telegraph. Read the original article at https://theconversation.com/religion-and-covid-19-methodists-and-church-of-england-followers-more-likely-to-have-been-vaccinated-than-muslims-and-pentecostals-228539 .There are many factors which affect how successfully a vaccine is rolled out. One of these is the public health communication strategy. Surprisingly, a key factor in determining the success of these strategies is religion. While some religious groups were keen to be vaccinated against COVID-19, others were much more hesitant.Economic & Social Research Council (COVID-19: Explaining the variance in people's trust and policy compliance); Brunel University London provides funding as a member of The Conversation UK
Dialectical Images: Two Films from Chile, ...A Valparaíso (Joris Ivens, 1963) and The Settlers (Felipe Gálvez Haberle, 2023)
Get PDFThis article is a revised version of the keynote speech delivered by Prof. Dr. Michael Wayne, moderated by Ertan Tunç, at the 7th International Cinema and Philosophy Symposium, organized by SineFilozofi Journal and the Cinema and Philosophy Association, held at CerModern in Ankara on December 7-8, 2024.
Bu makale, 7-8 Aralık 2024 tarihlerinde SineFilozofi Dergisi ile Sinema ve Felsefe Derneği tarafından Ankara Cer-Modern’de gerçekleştirilen 7. Uluslararası Sinema ve Felsefe Sempozyumu’nda Prof. Dr. Michael Wayne’in Ertan Tunç moderatörlüğünde yaptığı davetli konuşmanın gözden geçirilmiş versiyonudur.In this talk I trace the concept of the dialectical image back to Kant’s Critique of Judgment and his project which, retrospectively can be understood, against the dominant (bourgeois) interpretation, as a cognitive-affective illumination against reified concepts of the social order. Later, Marx’s critique of commodity fetishism laid the basis of modern critical thought with his dialectical analysis of the relations between forms of appearance and beneath them undisclosed social relations that contradict those appearance forms. With the expansion of the culture industry in the late nineteenth and twentieth century, the image as reified, frozen, undialectical hieroglyph and as a means for provoking and stimulating dialectical responses fascinated thinkers such as Kracauer and Benjamin and practitioners such as Eisenstein and Brecht. In the second part of my talk, I want to sketch out how we can apply the concept of the dialectical image (which includes its relationship to sound) in film. I will take two Chilean films, ...A Valparaiso (Joris Ivens1962) and The Settlers (Felipe Gálvez Haberle 2023) as examples. I will focus more on the recent film, because it is narrative based, and I want to show how narrative and dialectical structures are not necessarily opposed
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! 👇