- PublicationDesign Research Insights on Text Mining Analysis: Establishing the Most Used and Trends in Keywords of Design Research Journals(2022)Design research topics attract exponentially more attention and consideration among researchers. This study is the first research article that endeavors to analyze selected design research publications using an advanced approach called “text mining”. This approach speculates its results depending on the existence of a research term (i.e., keywords), which can be more robust than other methods/approaches that rely on contextual data or authors’ perspectives. The main aim of this research paper is to expand knowledge and familiarity with design research and explore future research directions by addressing the gaps in the literature; relying on the literature review, it can be stated that the research area in the design domain still not built-up a theory, which can unify the field. In general, text mining with these features allows increased validity and generalization as compared to other approaches in the literature. We used a text mining technique to collect data and analyzed 3553 articles collected in 10 journals using 17,487 keywords. New topics were investigated in the domain of design concepts, which included attracting researchers, practitioners, and journal editorial boards. Such issues as co-innovation, ethical design, social practice design, conceptual thinking, collaborative design, creativity, and generative methods and tools were subject to additional research. On the other hand, researchers pursued topics such as collaborative design, human-centered design, interdisciplinary design, design education, participatory design, design practice, collaborative design, design development, collaboration, design theories, design administration, and service/product design areas. The key categories investigated and reported in this paper helped in determining what fields are flourishing and what fields are eroding.
- PublicationLate Pleistocene-Holocene fluvial records of the Wadi Dishshah: hydro-climatic and archaeological implications (Southern piedmont of the Hajar Mountains, Oman)(2022)In Oman, quaternary climatic fluctuations alternated between humid and arid periods. Humid periods are a key component in landscape evolution and the history of early human-environment interactions, as they allowed for less-restrictive arid conditions by triggering increasing rainfall and fluvio-lacustrine activity. Fluvial archives are of great interest for understanding hydrosystems’ local responses to quaternary regional climatic fluctuations. For the end of the Pleistocene and the Holocene, little data are available in Northern Oman to examine this topic and to compare it with archaeological site distribution and subsistence strategies, in particular with regard to water resources. Here, we will present fluvial records from a small wadi called Wadi Dishshah, located in the southern part of the Hajar Mountains’ piedmont, near the Salakh Arch area. The study of the Wadi Dishshah relies on topographic surveys (aerial survey with drone), geomorphological mapping, morphostratigraphic analyses of natural and excavated sections, malacological analyses and age-dating using OSL and radiocarbon methods. Three phases of aggradation have been identified: the first one between 26,500 cal. BP and 11,300 cal. BP, a second between 6,200 cal. BP and 5,500 cal. BP and a late one around 2,800 cal. BP. The fluvial records from the Wadi Dishshah and its hydro-climatic significance are compared to the distribution of archaeological sites from the Salakh Arch area to discuss the relations between settlement strategies and surface flows. This work is the first case study of late Pleistocene – Holocene alluvial formations in this region of Oman.
- PublicationFirst‐Order River Delta Morphology Is Explained by the Sediment Flux Balance From Rivers, Waves, and Tides(2022)We present a novel quantitative test of a 50-year-old hypothesis which asserts that river delta morphology is determined by the balance between river and marine influence. We define three metrics to capture the first-order morphology of deltas (shoreline roughness, number of distributary channel mouths, and presence/absence of spits), and use a recently developed sediment flux framework to quantify the river-marine influence. Through analysis of simulated and field deltas we quantitatively demonstrate the relationship between sediment flux balance and delta morphology and show that the flux balance accounts for at least 35% of the variance in the number of distributary channel mouths and 42% of the variance in the shoreline roughness for real-world and simulated deltas. We identify a tipping point in the flux balance where wave influence halts distributary channel formation and show how this explains morphological transitions in real world deltas.
- PublicationVehicular Environment Identification Based on Channel State Information and Deep Learning(2022)This paper presents a novel vehicular environment identification approach based on deep learning. It consists of exploiting the vehicular wireless channel characteristics in the form of Channel State Information (CSI) in the receiver side of a connected vehicle in order to identify the environment type in which the vehicle is driving, without any need to implement specific sensors such as cameras or radars. We consider environment identification as a classification problem, and propose a new convolutional neural network (CNN) architecture to deal with it. The estimated CSI is used as the input feature to train the model. To perform the identification process, the model is targeted for implementation in an autonomous vehicle connected to a vehicular network (VN). The proposed model is extensively evaluated, showing that it can reliably recognize the surrounding environment with high accuracy (96.48%). Our model is compared to related approaches and state-ofthe-art classification architectures. The experiments show that our proposed model yields favorable performance compared to all other considered methods.
- PublicationA Novel Big Data Classification Technique for Healthcare Application Using Support Vector Machine, Random Forest and J48(2023)In this study, the possibility of using and applying the capabilities of artificial intelligence (AI) and machine learning (ML) to increase the effectiveness of Internet of Things (IoT) and big data in developing a system that supports decision makers in the medical fields was studied. This was done by studying the performance of three well-known classification algorithms Random Forest Classifier (RFC), Support Vector Machine (SVM), and Decision Tree-J48 (J48), to predict the probability of heart attack. The performance of the algorithms for accuracy was evaluated using the Healthcare (heart attack possibility) dataset, freely available on kagle. The data was divided into three categories consisting of (303, 909, 1808) instances which were analyzed on the WEKA platform. The results showed that the RFC was the best performer.