Research Title: Enhancing flat slab design: machine learning and metaheuristic approaches to predict punching shear strength

Author: Sawsan Mohammad Alkhawaldeh, Published Year: 2023

Asian Journal of Civil Engineering, 2023

Faculty: Engineering and Technology

Abstract: This study aimed to enhance the design of flat slabs by utilizing the capabilities of machine learning and metaheuristic optimization. Besides, the main aim of this study was to develop a robust predictive model for estimating punching shear strength, which is a crucial factor in the design of flat slabs and an attempt to enhance the safety and efficiency of construction practices. The study employed a carefully selected dataset from The American Concrete Institute Committee 445C, encompassing experimental findings about flat slabs. The Light Gradient Boosting Machine (LGBM) was the fundamental prediction model. The Locust Swarm Algorithm (LSA) was integrated to optimize the model's parameters and configurations to improve its performance. The combination of LGBM and LSA substantially improved prediction precision, accompanied by a noteworthy decrease in RMSE, MAE, and MAPE metrics. The hybrid approach performed better than the standalone LGBM model and traditional approaches. The findings highlight the possibility of combining machine learning and metaheuristic optimization techniques in structural engineering. The convergence of these factors can fundamentally change the approach to decision-making in flat slab designs, focusing on utilizing data-driven methods. This can significantly transform safety protocols and enhance the overall efficiency of construction processes. Moreover, the results have significant implications for the broader domain of civil engineering, indicating the advent of a novel era characterized by evidence-based and highly efficient design practices. The culmination of the research involves promoting the utilization of the LGBM-LSA hybrid as a robust method for predicting punching shear strength in flat slabs. The integration of machine learning and metaheuristic optimization in this study establishes a foundation for potential advancements in structural engineering.

Keywords: Flat slab design · Punching shear strength · Machine learning · LGBM · Metaheuristic optimization · Locust swarm algorithm

Research Title: Hybrid RNN and metaheuristic approach for modeling and optimization of seismic behavior in thin‑walled rectangular hollow bridge piers

Author: Sawsan Mohammad Alkhawaldeh, Published Year: 2023

Asian Journal of Civil Engineering, 2023

Faculty: Engineering and Technology

Abstract: In seismic structural engineering, there is a significant issue in comprehending the behavior of thin-walled rectangular hollow bridge piers within the context of dynamic phenomena. This research aimed to investigate a complex behavior using recurrent neural networks (RNNs) in conjunction with metaheuristic algorithms, namely the charged system search (CSS) and the black hole algorithm (BHA), to optimize the analysis. The approach used in this study included a rigorous process of data preprocessing to enhance the quality of seismic datasets and the development of an RNN model optimized utilizing the metaheuristics above. The results of the study were significant. The combined use of the RNN-CSS and RNN-BHA models showed enhanced prediction capacities compared to solo RNNs, thereby emphasizing the effectiveness of integrating neural networks with global optimization approaches. In addition, the convergence, diversity, search space, and sensitivity studies provided further insights into the modeling technique’s stability, comprehensiveness, and dependability. In summary, our study signifies a novel shift in seismic structural modeling, emphasizing the prospects of using multidisciplinary approaches to forecast and comprehend the hysteresis characteristics of bridge piers subjected to seismic stresses.

Keywords: Seismic structural engineering · Recurrent neural networks (RNNs) · Charged system search (CSS) · Black hole algorithm (BHA) · Thin-walled rectangular hollow bridge piers · Metaheuristic optimization

Research Title: Design, Synthesis, Anticancer Screening and Molecular Modelling Studies of Novel Thiazoles

Author: Mohammad Jamal Abu Nuwr, Published Year: 2023

ChemistrySelect, 8

Faculty: Pharmacy

Abstract: A library of novel benzamide-thiazolyl-chalcone compounds was synthesized via Claisen-Schmidt reaction and spectroscopically characterized using FT-IR, 1H-NMR, 13C-NMR, LC–MS, and HR-MS. The synthesized compounds were biologically evaluated using in vitro MTT assay against different cancer cell lines (MCF-7, MDA-AMB-231, Caco-2, A549, and H1299). Amongst all compounds screened, (E)-N-(5-(3-(4-hydroxyphenyl)-3-oxoprop-1-en-1-yl)thiazol-2-yl)benzamide and (E)-N-(5-(3-(4-methoxyphenyl)-3-oxoprop-1-en-1-yl)thiazol-2-yl)benzamide showed potent anti-proliferation against breast (MCF-7) and colon cancer cell lines reaching 88.56 % and 84.36 % with IC50 values of 44.00 μM and 58.88 μM, respectively. All synthesized compounds exhibited no significant cytotoxicity to normal cells. Structure-activity relationship studies demonstrated the effect of electron-donating and electron-withdrawing groups on the anticancer activity of the molecules under investigation. This was also corroborated by theoretical DFT studies. Thus, these molecules may serve as potential lead candidates for further development of novel anticancer agents against breast and colon cancers.

Keywords: Anti-cancer · Chalcone · DFT · Molecular modelling · Thiazole

Research Title: Secure Protocols in VANETs: Availability Considerations

Author: Maram Bani Younes, Published Year: 2023

The 14th International Conference on Information and Communication Systems, Irbid

Faculty: Information Technology

Abstract: The vehicular network technology provides several intelligent applications over road networks. These applications are classified into safety, efficiency, and infotainment based on their functionalities. Several studies have been proposed recently in this field where advanced and intelligent applications are developed for connected vehicles. This field of research seeks intensive and dedicated studies to enhance the confidentiality, integrity, and availability of the connecting network. In this work, we investigate the availability feature for vehicular network applications. First, we define the denial of service (DoS) attack and its different types. Then, we explore several possible DoS scenarios in the vehicular network environment. After that, we study the previous secure protocols that have generally considered and tackled the DoS attack on the vehicular network. Finally, we remark on the performance and functionalities of previous studies illustrating the main limitations and required work in this field. Thus we recommend future required work and studies in this field.

Keywords: Availability, DoS, VANETs, Applications

Research Title: Design, Synthesis, Anticancer Screening and Molecular Modelling Studies of Novel Thiazoles

Author: Adnan Dahadha, Published Year: 2023

Faculty: Science

Abstract: A library of 25 novel thiazole-chalcone hybrids were designed and synthesized. Anticancer activity was conducted using breast (MCF-7 and MDA-AMB-231), colon (Caco-2) and lung (A549 and H1299) cancer cell lines. Compounds (E)-N-(5-(3-(4-hydroxyphenyl)-3-oxoprop-1-en-1-yl)thiazol-2-yl)benzamide and (E)-N-(5-(3-(4-methoxyphenyl)-3-oxoprop-1-en-1-yl)thiazol-2-yl)benzamide showed 88.56 % and 84.36 % inhibition and IC50 values of 44.00 μM, 58.88 μM against MCF-7 cell line, respectively. Compounds (E)-N-(5-(3-(4-hydroxyphenyl)-3-oxoprop-1-en-1-yl)thiazol-2-yl)benzamide and (E)-N-(5-(3-oxo-3-phenylprop-1-en-1-yl)thiazol-2-yl)benzamide exhibited 75.83 % and 72.91 % inhibition against MDA-AMB-231 cell line.

Keywords: Anticancer activity, Thiazoles

Research Title: Building Trust in Fintech: An Analysis of Ethical and Privacy Considerations in the Intersection of Big Data, AI, and Customer Trust

Author: Hassan Hamad Aldboush, Published Year: 2023

International Journal o f Financial Studies, 11: 90.

Faculty: Business

Abstract: Abstract: This research paper explores the ethical considerations in using financial technology (fintech), focusing on big data, artificial intelligence (AI), and privacy. Using a systematic literature review methodology, the study identifies ethical and privacy issues related to fintech, including bias, discrimination, privacy, transparency, justice, ownership, and control. The findings emphasize the importance of safeguarding customer data, complying with data protection laws, and promoting corporate digital responsibility. The study provides practical suggestions for companies, including the use of encryption techniques, transparency regarding data collection and usage, the provision of customer opt-out options, and the training of staff on data-protection policies. However, the study is limited by its exclusion of non-English-language studies and the need for additional resources to deepen the findings. To overcome these limitations, future research could expand existing knowledge and collect more comprehensive data to better understand the complex issues examined.

Keywords: Keywords: fintech; big-data analytics; artificial intelligence (AI); data security and privacy; corporate digital responsibility (CDR); customer trust; ethical considerations

Research Title: Towards Automated Goal Model Generation from UML Use Case and Swimlane Diagrams

Author: Said Ahmad Ammar Ghoul, Published Year: 2023

International Journal Computer Aapplications, , Vol. 30, No. 2,

Faculty: Information Technology

Abstract: The Goal Model of software is one of the important concepts in the goal-based requirements engineering. It helps in specifying the software goals and the relationships between them. Several research works were conducted to generate Goal Model of software from its requirements documents. However, the generated Goal Models merge behavior and soft goals into a single model unit. This merging leads to tangled and complex generated Goal Models. Therefore, the maintenance of these models is hard and costly. The work presented in this paper proposes an approach splitting the generated Goal Model into three separated concerns (aspects) models (behavior, soft, and constraints) that facilitate its evolution and maintenance. The proposed approach is semi-automated, taking UML use case and Swimlane diagrams as inputs and generating a separated aspects model GM as output. The separation of Goal Model aspects led to adding new required information in input requirements specification documents. The feasibility of the proposed approach was validated on a concrete business case (Philadelphia University Quality Assurance Agenda). Its implementation was demonstrated through processes programming with algorithms and UML. Its contribution was demonstrated through its comparison with similar works. According to the observed results, this approach could be valuable in any goal-oriented requirements engineering application.

Keywords: Goal model (GM), behavior goal, soft goal, unified modeling language (UML), UML use case diagram, UML swim lane diagram, goal model generation, goal model maintenance, separation of concerns, aspects programming

Research Title: Toward Automated Feature Model Generation from UML Use Case and Class Diagrams

Author: Said Ahmad Ammar Ghoul, Published Year: 2023

International Journal Computer Aapplications, , Vol. 30, No. 2,

Faculty: Information Technology

Abstract: The Feature Model is one of the most important concepts in the Software Product Line development process, as it helps to represent commonalities and variabilities between software products. Several research works are currently conducted to generate feature model of a software product line from its requirements document. However, the generated feature models do not include data and actor aspects which are essential parts of any software product line. The work presented in this paper proposes an approach handling this insufficiency. This approach defines a methodology toward automated feature model generation from requirement documents limited to Unified Modelling Language use case and class diagrams. The use case is specified with a use case description template enhanced with information required for feature model. The class diagram is specified in a textual form obtained by translating the diagram into java programming language. The target generated feature model is enhanced by the introduction of data and actor concepts. An evaluating real case study (quality assurance at Philadelphia University), was used to evaluate the feasibility of the proposed approach and obtained results show its practical benefits.

Keywords: Feature model (FM), software production line (SPL), unified modelling language (UML) use case, UML class diagram, feature model generation approach

Research Title: BLIND COLOR USER INTERFACE REQUIREMENTS: A UML SPECIFICATION

Author: Said Ahmad Ammar Ghoul, Published Year: 2023

AIP, 2979, Issue 1

Faculty: Information Technology

Abstract: Color blindness, or color vision deficiency, is an inability to see the difference between certain colors. Men are more likely to be born with color blindness. Most people with color blindness can't distinguish between certain shades of red and green. Less commonly, people with color blindness can't distinguish between shades of blue and yellow. For that it’s a significant barrier to effective computer use. Several research works dealing with blind color user interfaces are currently conducted. However, requirements engineering for such interfaces is still missing. The work presented in this paper is part of an automated blind color user interface generation project which explores suitable input requirements document notations, simplifies the generating process, and outputs generated color blind user interface. It aims to define functional and nonfunctional requirements for powerful color blind user interface and their UML specification. Thus, different recent techniques that deal with these interfaces are selected according to their relevance and up to date, then they are analyzed, the lack of requirements engineering for these interface was stated, and a solution is proposed. It consists of requirements elicitation and specification with UML notations. A case study evaluating the proposed approach is presented. The obtained results confirm the power and positive impact of such requirements engineering on researches in this domain.

Keywords: color vision deficiency (CVD), color blindness requirements, color blindness user interface

Research Title: Automatically generated goal models from requirements: Toward an enhanced formalism.

Author: Said Ahmad Ammar Ghoul, Published Year: 2023

AIP , issue , 2979, Issue 1

Faculty: Information Technology

Abstract: The Goal Model (GM) of software is one of the important concepts in the software requirements specification, as it helps in representing the software goals and the relationships between them. Several research works were conducted to generate GM of software from its requirements document. However, the generated GMs merge behavior and soft goals aspects. This merging leaded to tangled and complex generated GMs that are difficult to maintain. The work presented in this paper proposes a GM formalism handling this insufficiency. It is part of an automated GM generating project where it focuses on the target GM. Thus, in this work, different recent techniques that automatically generate GMs are analyzed, their merging, tangling, and complexity of maintenance were stated, and a GM formalism facing these challenges is proposed. It separates goals clearly into behavior goals, soft goals, and constraints aspects. An evaluating case study, which is a partial concrete and real word domain requirements specification, is presented. This separation of concerns leaded to valuable results: complexity reduction, improvement of the understandability, simplicity of maintenance, and raising up the expressiveness power.

Keywords: Goal Model (GM), Behavior goal, Soft goal, Requirement specification, Automated goal model generation, Aspect separation, maintenance