Article
Computer Science, Artificial Intelligence
Ramsha Saeed, Hammad Afzal, Sadaf Abdul Rauf, Naima Iltaf
Summary: Continuous proliferation of hate speech on social media has become a major concern for researchers, who emphasize the importance of detecting and classifying its severity. This study focuses on detecting offensive and hate speech in the Urdu language, specifically targeting religion, racism, and national origin. The severity of hate speech is categorized into symbolization, insult, and attribution. An annotated corpus of over 20,000 tweets is collected and various models are applied to achieve high F-scores in hate speech detection, showing promising results for further investigation.
ACM TRANSACTIONS ON ASIAN AND LOW-RESOURCE LANGUAGE INFORMATION PROCESSING
(2023)
Article
Green & Sustainable Science & Technology
Gulsum Alicioglu, Bo Sun, Shen Shyang Ho
Summary: Traffic data analysis is essential for preventing fatal accidents by providing insights into the factors and patterns causing accidents. This study proposes a prediction model using various machine learning techniques and a negative data generator to address data bias, with neural networks outperforming other models in predicting injury severity.
Article
Green & Sustainable Science & Technology
Istiak Ahmad, Fahad Alqurashi, Ehab Abozinadah, Rashid Mehmood
Summary: In this paper, the concept of deep journalism and the tool DeepJournal (Version 1.0) are introduced, which utilizes data-driven deep learning to discover and analyze multi-perspective information for solving problems in smart cities and societies. By studying the transportation sector as a case study, it is found that there are important problems and gaps that industry and academia tend to overlook.
Article
Multidisciplinary Sciences
Yi Zou
Summary: Inspired by recent popular online business practices, this paper investigates the problem of reverse advance selling (RAS), a flipped procedure compared to traditional advance selling. The study considers competition and information asymmetry in the market and discusses their impact on decisions in reverse advance selling. Two models are proposed to evaluate the benefits of RAS and determine the optimal pricing and ordering policies for retailers in a competitive environment. The findings provide insights on the impact of market share, online reviews, and waiting time, and highlight the advantages of RAS in uncertain situations and the importance of updating review information.
Article
Economics
Sureyya Ozogur Akyuz, Birsen Eygi Erdogan, Ozlem Yildiz, Pinar Karadayi Atas
Summary: This study proposes a dynamic pricing procedure using a hybrid algorithm to estimate house prices. By combining different methods to deal with the heteroscedastic nature of housing data, the algorithm performs well in predicting housing prices.
COMPUTATIONAL ECONOMICS
(2023)
Article
Economics
Rounak Sil, Unninarayanan Kurup, Ashima Goyal, Apoorva Singh, Rajendra Narayan Paramanik
Summary: Based on consecutive MPC meetings of the Indian central bank, we have developed two innovative measures of implicit dissent at both individual and group levels. By utilizing VADER sentiment analysis, we have examined the impact of these measures on anchoring growth and inflation forecasts in India. Empirical findings demonstrate that increased discordance among committee members enhances forecast accuracy, highlighting the importance of fostering an environment that embraces diverse opinions for better policy outcomes.
APPLIED ECONOMICS LETTERS
(2023)
Article
Green & Sustainable Science & Technology
Eman Alqahtani, Nourah Janbi, Sanaa Sharaf, Rashid Mehmood
Summary: Smart homes play a critical role in establishing smart living and enabling sustainable cities and societies. However, current research on smart homes mainly focuses on developing specific functions, such as security and ambiance management, while the understanding of families and their complex dynamics is still limited. This paper introduces a data-driven parameter discovery methodology and provides a comprehensive analysis of the families and homes landscape using academic literature and public opinions from social media. Through deep learning and natural language processing, the paper discovers 66 parameters and a multidimensional knowledge space, which can contribute to the development of community-specific policies, technologies, and solutions for enhancing families and homes, ultimately leading to empowered and sustainable societies globally.
Article
Environmental Studies
Zheng-Zheng Li, Chi-Wei Su, Tsangyao Chang, Oana-Ramona Lobont
Summary: This study verifies the existence of multiple bubbles in the steam coal market in China and identifies policy changes as the main driver. Bubbles originate in producing regions and spread to consumption areas, with demand-side factors driving the diffusion. The findings suggest reducing policy intervention in the coal industry for a market-oriented price mechanism and stability.
Article
Economics
Probal P. Ghosh
Summary: India heavily depends on imported crude for petroleum consumption, with diesel being a significant portion. Government subsidies and price controls aimed to reduce fiscal deficit, but the actual economic growth did not meet expectations.
Article
Computer Science, Information Systems
Ching Wai Yong, Kareen Teo, Belinda Pingguan Murphy, Yan Chai Hum, Yee Kai Tee, Kaijian Xia, Khin Wee Lai
Summary: Osteoarthritis is a common form of knee arthritis that causes significant disability and there is currently no known cure for it. Early identification is crucial for clinical interventions. A proposed ordinal regression module for neural networks showed significant improvements in predicting knee OA KL grade.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Environmental Studies
Yangzhe Cao, Brent Swallow, Feng Qiu
Summary: Open spaces provide a range of benefits in economic, social and environmental aspects, but population growth and changes in economic growth forms can lead to their conversion to developed uses. Different land-use policies can affect property values, but overlooking the impact of self-selection can result in biased policy assessments. The study finds that the introduction of a new development policy in Okotoks, Canada led to a decline in willingness to pay for open space and a reduction in residential real estate value, highlighting the importance of considering self-selection bias in policy evaluations.
Article
Economics
Boqiang Lin, Penghu Zhu
Summary: The study shows that policy cognition can reduce electricity consumption by about 42.25%, with price cognitive bias being a crucial way to achieve this effect. Especially for residents with low incomes, low electricity consumption, and over 50 years old, this energy-saving effect is more significant.
Article
Green & Sustainable Science & Technology
Yasir Ali, Ahmed Raza, Sidra Iqbal, Azhar Abbas Khan, Hafiz Muhammad Aatif, Zeshan Hassan, Ch Muhammad Shahid Hanif, Hayssam M. Ali, Walid F. A. Mosa, Iqra Mubeen, Lidia Sas-Paszt
Summary: In this study, disease predictive models were developed to forecast the severity and yield loss of wheat leaf rust. The models showed high accuracy in predicting the two variables and can be used by farmers for disease management decisions. The research highlights the importance of using forecasting models for effective use of fungicides and limiting crop yield losses.
Article
Green & Sustainable Science & Technology
Maowei Chen, Lele Zhou, Sangho Choo, Hyangsook Lee
Summary: This study analyzes the severity of truck traffic accidents in Korea, identifying the driver, environmental, and traffic condition factors that contribute to the severity and proposing practical truck safety solutions. The study also finds that truck traffic accidents in logistics influencing areas are more likely to be serious.
Article
Green & Sustainable Science & Technology
Raniah Alsahafi, Ahmed Alzahrani, Rashid Mehmood
Summary: Global events are exposing the fragility of the tourism industry and its impact on the global economy. Prior to the COVID-19 pandemic, tourism contributed significantly to global GDP and employment, but experienced a decline due to the pandemic. Sustainable and smart tourism requires collaboration and comprehensive understanding to drive responsible and innovative growth in the sector.
Article
Mathematical & Computational Biology
Fengqing (Zoe) Zhang, Don Hong
STATISTICS IN MEDICINE
(2011)
Article
Economics
Donglin Wang, Don Hong, Qiang Wu
Summary: Performance analysis of using deep neural network for loan rate prediction
COMPUTATIONAL ECONOMICS
(2023)
Article
Biochemical Research Methods
Donglin Wang, Don Hong, Qiang Wu
Summary: In this study, two novel deep learning approaches for ADHD classification based on functional magnetic resonance imaging were proposed. Both methods outperform traditional classification methods and have shown great potential in clinical applications.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2023)
Article
Automation & Control Systems
Zhengzheng Li, Jiancheng Zou, Peizhou Yan, Don Hong
Summary: The paper introduces a non-contact real-time monitoring algorithm for physiological parameters of drivers under ambient light conditions, using facial expression recognition and independent component separation to monitor the driver's physiological parameters, providing early warnings to help prevent traffic accidents.
INTELLIGENT AUTOMATION AND SOFT COMPUTING
(2021)
Article
Computer Science, Information Systems
Yuyang Sun, Peizhou Yan, Zhengzheng Li, Jiancheng Zou, Don Hong
CMC-COMPUTERS MATERIALS & CONTINUA
(2020)
Article
Computer Science, Information Systems
Jiancheng Zou, Zhengzheng Li, Zhijun Guo, Don Hong
CMC-COMPUTERS MATERIALS & CONTINUA
(2019)
Article
Mathematics, Interdisciplinary Applications
Jingsai Liang, Jiancheng Zou, Don Hong
FRONTIERS IN APPLIED MATHEMATICS AND STATISTICS
(2019)
Article
Economics
Bingjie Wang, Jinzhu Li
Summary: This paper focuses on the asymptotic behavior of a popular risk measure called the tail moment (TM). The study reveals precise asymptotic results for the TM under scenarios where individual risks are mutually independent or have a specific dependence structure. Furthermore, the article provides an analysis of the relative errors between the asymptotic results and the exact values.
INSURANCE MATHEMATICS & ECONOMICS
(2024)
Article
Economics
Guangyuan Gao
Summary: This article proposes a new method for fitting the Tweedie model, which uses the EM algorithm to address heterogeneous dispersion and estimate the power variance parameter.
INSURANCE MATHEMATICS & ECONOMICS
(2024)
Article
Economics
Anna Rita Bacinello, Rosario Maggistro, Ivan Zoccolan
Summary: In this paper, a model is proposed for pricing GLWB variable annuities under a stochastic mortality framework. The contract value is defined through an optimization problem solved by using dynamic programming. The authors prove the validity of the bang-bang condition for the withdrawal strategies of the model using backward induction. Extensive numerical examples are presented, comparing the results for different parameters and policyholder behaviours.
INSURANCE MATHEMATICS & ECONOMICS
(2024)
Article
Economics
Sascha Gunther, Peter Hieber
Summary: The financial return of equity-indexed annuities depends on an underlying fund or investment portfolio complemented by an investment guarantee. This study introduces a novel scenario-matrix method for valuation and risk management, specifically for the cliquet-style or ratchet-type guarantee. Numerical tests show that this method outperforms existing approaches in terms of computation time and accuracy.
INSURANCE MATHEMATICS & ECONOMICS
(2024)