Complex Real-world Problems Research Articles

Knowledge Graph-Based Convolutional Network Coupled With Sentiment Analysis Towards Enhanced Drug Recommendation.

Recommending appropriate drugs to patients based on their history and symptoms is a complex real-world problem. Knowing whether a drug is useful without its consumption by a variety of people followed by proper evaluation is a challenge. Modern-day recommender systems can assist in this provided they receive large data to learn. Public reviews on various drugs are available for knowledge sharing. These reviews assist in recommending the best and most appropriate option to the user. The explicit feedback underpins the entire recommender system. This work develops a novel knowledge graph-based convolutional network for recommending drugs. The knowledge graph is coupled with sentiment analysis extracted from the public reviews on drugs to enhance drug recommendations. For each drug that has been used previously, sentiments have been analyzed to determine which one has the most effective reviews. The knowledge graph effectively captures user-item relatedness by mining its associated attributes. Experiments are performed on public benchmarks and a comparison is made with closely related state-of-the-art works. Based on the obtained results, the current work performs better than the past contributions by achieving up to 98.7% Area Under Curve (AUC) score.

Redesigning of Signals and Systems Course to Emphasize Problem/Project Based and Experiential Learning

prime trait of engineering education which persistently challenges to tide over classroom and real-world scenario is its cornerstone on fortifying future technocrats to design and implement various systems of engineering. This paper introduces the experimental study of signals and systems course in an under graduate(UG) program by using problem/project-based learning at Anurag University, which makes the learners to acquire iterative and collaborative skills to solve real world complex problems in practical approach subjects. The major objective of this framework is to enrich the conventional course of study with value creation, responsive content, practices and larks providing a competitive scope to learners. The specified approach do enhances the learning spirit among the learners by doing the things which intern results in experiential learning. The problem/project based learning enactment frame and constructive alignment of course outcomes, teaching/learning methods and assessment are specified in the paper. Keywords: Constructive alignment, Engineering education, Problem/Project-based learning, Signals and Systems

A robust computational study for assessing the dynamics and control of emerging zoonotic viral infection with a case study: A novel epidemic modeling approach

Fractional calculus and fractal theory remain significant tools in modeling complex real-world problems in biology and life science. In this study, we formulated a compartmental mathematical model using the Caputo fractional and fractal–fractional operators to study the dynamics and transmission of Nipah virus infection. Initially, the model is developed by a system of seven nonlinear ordinary differential equations that govern the dynamics of viral concentration, the flying fox, and the human populations. Furthermore, the model is restructured using more general modeling approaches based on fractional calculus and fractal theory to gain valuable insights into the dynamics of Nipah virus transmission. The necessary properties of the model, such as uniqueness and existence in both cases, were investigated, and possible equilibrium points with their existence were presented. The model parameters are estimated on the basis of the clinical epidemiology of the Nipah outbreak in Bangladesh, one of the most affected regions. The stability of the fractional model is studied by applying the Ulam–Hyers and Ulam–Hyers–Rassias stability conditions. Moreover, computational schemes for the model in fractional and fractal–fractional cases are developed using interpolation techniques. Finally, a detailed simulation was presented to validate the theoretical findings. We affirm that the present findings will help researchers incorporate advanced computational techniques in infectious disease modeling and control studies.

Explainability Enhanced Object Detection Transformer with Feature Disentanglement.

Explainability is a pivotal factor in determining whether a deep learning model can be authorized in critical applications. To enhance the explainability of models of end-to-end object DEtection with TRansformer (DETR), we introduce a disentanglement method that constrains the feature learning process, following a divide-and-conquer decoupling paradigm, similar to how people understand complex real-world problems. We first demonstrate the entangled property of the features between the extractor and detector and find that the regression function is a key factor contributing to the deterioration of disentangled feature activation. These highly entangled features always activate the local characteristics, making it difficult to cover the semantic information of an object, which also reduces the interpretability of single-backbone object detection models. Thus, an Explainability Enhanced object detection Transformer with feature Disentanglement (DETD) model is proposed, in which the Tensor Singular Value Decomposition (T-SVD) is used to produce feature bases and the Batch averaged Feature Spectral Penalization (BFSP) loss is introduced to constrain the disentanglement of the feature and balance the semantic activation. The proposed method is applied across three prominent backbones, two DETR variants, and a CNN based model. By combining two optimization techniques, extensive experiments on two datasets consistently demonstrate that the DETD model outperforms the counterpart in terms of object detection performance and feature disentanglement. The Grad-CAM visualizations demonstrate the enhancement of feature learning explainability in the disentanglement view.

A review of reinforcement learning based hyper-heuristics.

The reinforcement learning based hyper-heuristics (RL-HH) is a popular trend in the field of optimization. RL-HH combines the global search ability of hyper-heuristics (HH) with the learning ability of reinforcement learning (RL). This synergy allows the agent to dynamically adjust its own strategy, leading to a gradual optimization of the solution. Existing researches have shown the effectiveness of RL-HH in solving complex real-world problems. However, a comprehensive introduction and summary of the RL-HH field is still blank. This research reviews currently existing RL-HHs and presents a general framework for RL-HHs. This article categorizes the type of algorithms into two categories: value-based reinforcement learning hyper-heuristics and policy-based reinforcement learning hyper-heuristics. Typical algorithms in each category are summarized and described in detail. Finally, the shortcomings in existing researches on RL-HH and future research directions are discussed.

A Survey On Non – linear Optimization And Global Optimization Methods

Non-linear optimization and global optimization methods are crucial in solving complex real-world problems where the objective functions and constraints are not linear. Non-linear optimization involves finding the best solution in the presence of non-linear relationships among the variables. These problems arise in various fields such as engineering design, economics, machine learning, and operational research. The primary challenge in non-linear optimization is the potential existence of multiple local optima, which can make it difficult to identify the global optimum. Global optimization methods are designed to overcome this challenge by searching for the global optimum, which is the best possible solution among all local optima. These methods can be broadly categorized into deterministic and stochastic approaches. Deterministic methods, such as branch and bound, interval analysis, and cutting-plane methods, systematically explore the search space and provide guarantees of finding the global optimum or bounds on the global optimum. These methods are often rigorous but can be computationally expensive, especially for high-dimensional problems. On the other hand, stochastic methods, such as genetic algorithms, simulated annealing, and particle swarm optimization, use probabilistic rules to explore the search space. These methods are inspired by natural processes and heuristics, offering flexibility and often being more efficient for large-scale problems. While they do not guarantee finding the global optimum, they are widely used due to their ability to escape local optima and explore the search space effectively. Hybrid methods that combine deterministic and stochastic approaches have also gained popularity. These methods leverage the strengths of both approaches, enhancing the robustness and efficiency of the optimization process. For example, a hybrid algorithm might use a stochastic method to explore the search space broadly and then apply a deterministic method to refine the search around promising areas. Recent advancements in computational power and algorithmic development have significantly improved the efficiency and applicability of non-linear and global optimization methods. Techniques such as machine learning-based optimization and parallel computing have enabled the handling of more complex and larger-scale optimization problems. Moreover, these advancements have facilitated the integration of optimization methods into real-time and adaptive systems, further broadening their applications.

Computational Fluid Dynamics Analysis of 3D Wind Flow around Rhizophora Mangrove Tree and Drag Force Acting on the Wall of the Tree

Many natural disasters, such as cyclones, typhoons, mudslides, and tsunamis, are currently plaguing the world. These tragedies, which have destroyed the coastal settlements, claim a great number of lives each year. The characteristics and structure of the Mangrove tree have the significant impact on protecting the coast from heavy wind, floods, mudslides, and other kinds of natural disasters. This research work investigates how Mangroves can protect the area near the coast from a fluid dynamics point of view. Complex real world problems require intelligent systems that combine knowledge and techniques. Hence Computational Fluid Dynamics (CFD) technique namely the finite volume method is employed as a tool in this study. In this research, a model Rhizophora Mangrove tree from Pichavaram Mangrove forest, Tamilnadu, India is generated using Ansys Workbench design modeler. A computational domain is created around this tree to reduce the boundary effect. Grid-independent research was carried out using various mesh sizes to determine the best grid resolution for the analysis. An unstructured triangular mesh was generated for the simulation. The focus of this study is to determine the 3D flow around the tree in order to determine the wind velocity, as well as the coefficient of drag and drag force on the wall tree. This research helps to give a clear understanding of wind flow around the 3D Rhizophora Mangrove tree. The velocity profile, pressure distribution, drag force, and drag coefficient around the tree are captured, analysed, and presented in detail. The results show that the Rhizophora mangrove trees can significantly reduce the flow velocity of the wind and will be able to safeguard the coast and communities nearby from natural disasters.

Bridging the Gap Between Quantum Computing and Artificial Intelligence

Quantum machine learning (QML) holds the potential to transform various industries, yet its widespread adoption faces formidable challenges. This paper provides a condensed exploration of these challenges and opportunities. I delve into: • Hardware Limitations: Present quantum computers are constrained in qubit count and gate quality, posing obstacles to real-world QML implementation. We dissect the implications of these limitations on QML computations. • Error Correction: Quantum systems are prone to errors stemming from hardware noise and gate imperfections. We scrutinize the strategies to mitigate these errors and enhance QML accuracy. • Algorithm Development: The evolution of QML algorithms is examined, including computational complexity issues and adaptability to noisy quantum hardware. then pivot to recent developments in quantum hardware, which are not only addressing these challenges but also generating new prospects: • Addressing Challenges: Increased qubit counts and improved qubit quality empower quantum computers to tackle complex real-world problems, with reduced noise and enhanced scalability. • Contributing to Opportunities: Recent hardware advancements have catalyzed the creation of efficient quantum algorithms and the exploration of novel quantum applications in healthcare, finance, and materials science. • Creating Opportunities: Quantum computing, with its unique capabilities, has the potential to unlock scientific and engineering discoveries unattainable through classical computing, ushering in a new era of innovation across diverse industries. As quantum hardware continues to advance, these developments are poised to shape the future of quantum machine learning and its profound impact on society.

A novel optimization approach based on unstructured evolutionary game theory

Proposing new metaheuristic methods is crucial for continuous algorithmic improvement and the ability to effectively address increasingly complex real-world optimization problems. On the other hand, Evolutionary Game Theory analyzes how trough competition is possible to modify the strategies of individuals within a population in order to spread successful mechanisms and reduce or remove less successful strategies. This paper introduces a novel optimization approach based on the principles of evolutionary game theory. In the proposed method, all individuals are initialized using the Metropolis–Hasting technique, which sets the solutions at a starting point closer to the optimal or near-optimal regions of the problem. An original strategy is then assigned to each individual in the population. By considering the interactions and competition among different agents in the optimization problem, the approach modifies the strategies to improve search efficiency and find better solutions. To evaluate the performance of the proposed technique, it is compared with eight well-known metaheuristic algorithms using 30 benchmark functions. The proposed methodology demonstrated superiority in terms of solution quality, dimensionality, and convergence when compared to other approaches.

A novel CODAS approach based on Heronian Minkowski distance operator for T-spherical fuzzy multiple attribute group decision-making

T-spherical fuzzy sets (TSFSs) are more flexible and efficient tools to deal with ambiguous, uncertain and vague information in complex real-world decision-making problems than various extended intuitionistic fuzzy sets. This paper aims to develop a novel T-spherical fuzzy (TSF) Combinative Distance-Based ASsessment (CODAS) based on the Heronian Minkowski distance aggregation operator, this new method can capture interrelationship between input arguments. Some TSF weighted Heronian Minkowski distance (TSFWHMD) aggregation operators with generalization are developed based on Heronian mean and Minkowski-type distance, their properties are discussed as well as their families are analyzed. Furthermore, the TSF MAGDM methodology based on the improved CODAS is designed, where the Minkowski-type distance is used to define the TSF similarity for computing the expert weights and to construct the maximizing deviation method (MDM) for determining the attribute weights, respectively. The TSF ordered weighted Heronian Hamming distance (TSFOWHHD) and TSF ordered weighted Heronian Euclidean distance (TSFOWHED) operators derived from the TSFOWHMD operator are integrated into the CODAS method, which is an improved method for both measuring the deviation of the negative ideal solution from each alternative and capturing the correlation between attributes. Finally, the feasibility and practicality of developed methodology are illustrated with an example of CAE (Computer Aided Engineering) software selection for lithium-ion power battery (LiPB) design, sensitivity analysis and method comparisons are performed to elucidate the reliability and validity of the developed methodology.

Attraction–Repulsion Optimization Algorithm for Global Optimization Problems

In this study, a novel meta-heuristic search (MHS) algorithm for constrained global optimization problems is proposed. Since many algorithms aim to achieve well-balanced exploitation–exploration stages with often unsatisfactory results, in the approach introduced in this paper, Attraction–Repulsion Optimization Algorithm (AROA), the balance associated with attraction–repulsion phenomena that occur in nature is mimicked. AROA introduces a search strategy in which a candidate solution is moved in the search space depending on the quality of solutions in its neighborhood, as well as the best candidate. The candidates are managed by local search operators based on modified Brownian motion, trigonometric functions, randomly selected solutions, and a form of memory. Consequently, AROA exhibits a satisfactory exploitation–exploration balance exhibited by highly competitive performance. The introduced algorithm is experimentally compared with the state-of-the-art meta-heuristics on the CEC 2014, 2017, and 2020 test suites. The obtained results reveal the advantages of AROA over related algorithms and its suitability in solving complex real-world problems.

Insights of 6G and artificial intelligence-based internet-of vehicle towards communication

<p>The significance of internet-of-vehicle (IoV) is spontaneously increasing with exponentially rising demands towards transportation system and road safety. At present, there are various number of scientific approaches which is meant for leveraging the communication performance in IoV, but yet the problem still exists over multiple attributes e.g. resource management, privacy, security, and service offloading. Such problems are anticipated to be solved by 6G services that offers better communication capabilities compared to its prior version of 5G. At the same time, the quality of communication system can be enhanced by artificial intelligence (AI), which is capable of solving complex real-world problems. Therefore, this manuscript offers an insight towards strength and weakness of existing 6G based study model as well as AI-based solution in order to contribute towards highlighting an essential research gap that could directly offer better insight towards future planning towards improving communication in IoV.</p>

Taking pluralism seriously: a new perspective on evidence-based policy

Abstract Scientific policy advice in the face of complex real-world problems requires a maximally pluralistic knowledge base. However, integrating different types of knowledge from a variety of sources raises a series of highly challenging epistemic as well as practical problems. We propose an integrated model of pluralistic policy advice emphasizing the appropriate constitution and structure of pluralistic expert panels. Furthermore, we suggest that normative assumptions underlying expert advice should be made explicit to policymakers. The goal should not be to avoid normative biases, but rather to create transparency with respect to the unavoidable value judgements that influence the epistemic preferences of experts and to provide policymakers with evidence-based scenarios that range over a variety of value preferences.

The Role of Problem-Based Learning Approach in Teaching and Learning Physics: A Systematic Literature Review.

Problem-based learning (PBL) is a learner-centered method in which complex real-world problems are used to stimulate students thinking and problem-solving abilities during the teaching and learning process. This systematic review aimed to reveal the role of the PBL approach in teaching and learning physics. Relevant articles for the review were sourced from Scopus and Web of Science using keywords such as "problem-based learning" and "PBL in physics" education" as search terms. This search yielded 376 results. Thirty-six articles were included in the analysis after passing a crucial condition of empirically investigating the effect of PBL in teaching and learning physics. Only three of these articles did not show a positive effect; others have shown a positive lead of PBL towards improving academic achievement, attitude to learn physics, problem-solving, critical and creative thinking abilities, cooperative learning enhancement, mental model development, and science process skills attainment. Therefore, the review offers important pointers to various classroom environments and activities that ignite learners' thinking. Thus, it help policymakers to select and maintain the best methodology that promotes high students' academic achievement levels.

New chaos-integrated improved grey wolf optimization based models for automatic detection of depression in online social media and networks

Depression is a psychological effect of the modern lifestyle on people’s thoughts. It is a serious individual and social health problem due to the risk of suicide and loss of workforce, high chronicity, recurrence rates, and prevalence. Therefore, identification, prevention, treatment of depression, and determination of relapse risk factors are of great importance. Depression has traditionally been diagnosed using standardized scales that require clinical diagnoses or patients’ subjective responses. However, these classical techniques have some limitations such as cost, uncomfortability, subjectivity, and ineffectiveness. Social media data can be simply and efficiently used for depression detection because it allows instantaneous emotional expression and quick access to various information. Some machine learning-based methods are used for detecting the depression in online social media and networks. Nevertheless, these algorithms suffer from several drawbacks, including data sparsity, dimension explosion, restricted capacity for generalization, and low performance on imbalanced data sets. Furthermore, many machine learning methods work as black-box models, and the constructed depression detection models are not interpretable and explainable. Intelligent metaheuristic optimization algorithms are widely used for different types of complex real-world problems due to their simplicity and high performance. It is aimed to remove the limitations of studies on this problem by increasing the success rate and automatically selecting the relevant features and integrating the explainability. In this study, new chaos-integrated multi-objective optimization algorithms are proposed to increase efficiency. New improved Grey Wolf Optimization algorithms have been proposed by integrating Circle, Logistic, and Iterative chaotic maps into the improved Grey Wolf Optimization algorithm. It is aimed to increase the success rate by proposing a multi-objective fitness function that can optimize the accuracy and the number of features simultaneously. The proposed algorithms are compared with different types of popular supervised machine learning algorithms and current metaheuristic algorithms that are widely and successfully used in depression detection problems. Experimental results show that the proposed models outperform machine learning methods, as evidenced by examining results with accuracy, F-measure, MCC, sensitivity, and precision measures. An accuracy value of 100% was obtained from proposed algorithms. In addition, when the confusion matrices are examined, it is seen that they exhibit a successful distribution. Although it is a new research and application area for optimization theory, promising results have been obtained from the proposed models.

A Learning based Secure Routing Approach using Deep Reinforcement Learning in IoT Integrated Wireless Sensor Network

The usage of Wireless Sensor Network (WSN) is ubiquitous in nature. With the emergence of Internet of Things (IoT) technology and its unprecedented use cases, the role of sensor networks as part of IoT application became crucial. WSN became backbone of IoT to realize integration between physical and digital worlds and connectivity to Internet. However, IoT devices are resource constrained with limited computational capabilities. The entire network is distributed in nature and has increased complexity. Routing in such WSN integrated IoT network plays an important role in achieving meaningful communication among objects. In this context, it is indispensable to have more energy efficient routing method. Since the IoT integrated sensor network is highly complicated, it is very dynamic in nature. Thus routing decisions are also dynamic leading to much importance to routing in such use cases. With the emergence of Artificial Intelligence (AI), it became possible to solve complex real world problems through learning based approach which acquires desired intelligence prior to making decisions. In this paper we proposed a deep reinforcement learning based routing mechanism for energy efficient routing in WSN-IoT integrated application. We proposed novel algorithms for network setup, formation of clusters and routing. Our method adapts to network changes due to energy levels, mobility and makes learning based routing decisions. We enhanced the method further with security to ensure its Qualityof Service (QoS) in presence of attacks. Our simulation study using MATLAB has revealed that the proposed secure routing approach outperforms existing protocols.

Supporting scheduling decisions by using genetic algorithm based on tourists’ preferences

As the tourism industry switches from mass to personalized solutions, tourists’ preferences should be considered. This study aims to optimize tourists’ activity chains based on their preferences and 10 realistic parameters. A method using activity-based modeling and genetic algorithm is developed. The method handles flexible touristic activities with time window restrictions. Then scenarios including private vehicles (PV) and public transport (PT) are created. The results show high priority for security and provide significant reduction in journey times, particularly in case of flexible activities. The outcomes confirm that the proposed method provides an optimized solution to generate personalized travel itineraries that enhance tourists’ travel experience. To evaluate the performance of the algorithm, Wilcoxon Signed-Ranks Test is applied. The statistical results confirm that the performance of the GA significantly differs in the various scenarios. Travelers can benefit from the proposed approach since it offers a solution to complex real-world scheduling problems. In addition, travel patterns can be derived thus supporting the work of planners and decision-makers in the tourism industry.

Low-Dimensional Space Modeling-Based Differential Evolution for Large-Scale Global Optimization Problems

Large-Scale Global Optimization (LSGO) has been an active research field. Part of this interest is supported by its application to cutting-edge research such as Deep Learning, Big Data, and complex real-world problems such as image encryption, real-time traffic management, and more. However, the high dimensionality makes solving LSGO a significant challenge. Some recent research deal with the high dimensionality by mapping the optimization process to a reduced alternative space. Nonetheless, these works suffer from the changes in the search space topology and the loss of information caused by the dimensionality reduction. This paper proposes a hybrid metaheuristic, so-called LSMDE (Low-dimensional Space Modeling-based Differential Evolution), that uses the Singular Value Decomposition to build a low-dimensional search space from the features of candidate solutions generated by a new SHADE-based algorithm (GM-SHADE). GM-SHADE combines a Gaussian Mixture Model (GMM) and two specialized local algorithms: MTS-LS1 and L-BFGS-B, to promote a better exploration of the reduced search space. GMM mitigates the loss of information in mapping high-dimensional individuals to low-dimensional individuals. Furthermore, the proposal does not require prior knowledge of the search space topology, which makes it more flexible and adaptable to different LSGO problems. The results indicate that LSMDE is the most efficient method to deal with partially separable functions compared to other state-of-the-art algorithms and has the best overall performance in two of the three proposed experiments. Experimental results also show that the new approach achieves competitive results for non-separable and overlapping functions on the most recent test suite for LSGO problems.

Multi-Objective Deep Reinforcement Learning for Crowd Route Guidance Optimization

In this study, we propose an improved version of Pareto deep Q-network (PDQN), a multi-objective deep reinforcement learning method, and attempt to demonstrate its effectiveness in a real-world problem such as crowd route guidance strategy optimization. Overcrowding during crowd movement can sometimes lead to accidents; therefore, it is imperative to guide crowds to move safely and efficiently. Safety and efficiency are conflicting objectives, and how to dynamically determine guidance can be formulated as a multi-objective sequential decision-making problem. PDQN is suitable for solving these problems, but its applicability to complex real-world problems has not been fully verified. To apply PDQN to real problems, we propose to adjust the parameters of PDQN and improve the action selection criteria during learning. A toy problem and a crowd guidance problem using multi-agent crowd simulation are adopted to evaluate the performance of improved PDQN. Experimental results show that the improved PDQN can search for good strategies compared with the original PDQN and can obtain better strategies than simplified strategies.

The Role of Problem-Based Learning Approach in Teaching and Learning Physics: A Systematic Literature Review

Problem-based learning (PBL) is a learner-centered method in which complex real-world problems are used to stimulate students thinking and problem-solving abilities during the teaching and learning process. This systematic review aimed to reveal the role of the PBL approach in teaching and learning physics. Relevant articles for the review were sourced from Scopus and Web of Science using keywords such as “problem-based learning” and “PBL in physics” education” as search terms. This search yielded 376 results. Thirty-six articles were included in the analysis after passing a crucial condition of empirically investigating the effect of PBL in teaching and learning physics. Only three of these articles did not show a positive effect; others have shown a positive lead of PBL towards improving academic achievement, attitude to learn physics, problem-solving, critical and creative thinking abilities, cooperative learning enhancement, mental model development, and science process skills attainment. Therefore, the review offers important pointers to various classroom environments and activities that ignite learners’ thinking. Thus, it help policymakers to select and maintain the best methodology that promotes high students’ academic achievement levels.