Dell Inc.

Automatic Speech Recognition (ASR) systems have been widely used as a practical method of interaction between humans and devices. They are typically employed to enhance the accessibility of devices and to improve the security of systems, among other purposes. However, the design of speech‐based systems imposes many challenges due to their particularities. Currently, the majority of ASR systems is based on the Hidden Markov Model (HMM), and, more recently, on Convolutional Neural Networks (CNN). The present research evaluates the performance of Hidden Markov Model (HMM) and Convolutional Neural Network (CNN) algorithms in speech recognition and proposes a novel hybrid approach that combines both methods. The study assesses various performance metrics, including accuracy, precision, recall, F1‐score, response time, and computational cost. The experimental tests show that the integration between HMM and CNN increased the accuracy by 6% and 8% when compared to HMM and CNN isolated, respectively, in accordance with results presented in previous papers. However, the results of the ANOVA test revealed that the difference in question is not statistically significant, and the HMM‐only approach still being an interesting option for embedded systems due to its lesser demanded computational effort.

Advanced sensor technologies are revolutionizing the landscape of green transportation systems, offering unprecedented opportunities for enhancing sustainability, efficiency, and environmental stewardship in the transport sector. This chapter delves into the myriad of sensor applications that are reshaping how we conceptualize and implement eco-friendly transportation solutions. From environmental monitoring to vehicle optimization and intelligent traffic management, sensors are at the forefront of the green transportation revolution. By examining cutting-edge sensor technologies, their integration into transportation infrastructure, and their impact on reducing emissions and improving energy efficiency, this chapter provides a comprehensive overview of the current state and future potential of sensor-driven green transportation. Through a combination of theoretical analysis, experimental data, and real-world case studies, we explore how these technologies are not only addressing current environmental challenges but also paving the way for a more sustainable and intelligent transportation ecosystem.

Polycystic ovary syndrome (PCOS) is a common yet complex endocrine disorder in women of reproductive age. It is associated with several metabolic complications necessitating personalized management approaches. However, current strategies fail to capture the heterogeneity in PCOS phenotypes. Nutritional therapy tailored to distinct metabolic profiles underlying PCOS subtypes has immense yet untapped potential. This chapter proposes an artificial intelligence (AI) based precision nutrition model to provide customized dietary recommendations for PCOS by systematically assimilating multi-domain data, utilizing explainable AI to decipher pathology, matching dietary components to outcomes via machine learning algorithms and continually updating recommendations. It also delineates opportunities and challenges around advancing adoption of such AI-enabled personalized nutrition prescriptions to transform care. In nutshell, AI-enabled, tailored nutritional therapy carries immense potential to transform PCOS care from one-size-fits-all strategies to individualized prescriptions effectively addressing the intricate metabolic heterogeneity underlying this complex disorder. The long-term impact on outcomes cannot be overstated given the alarming burden inflicted by this highly disruptive syndrome on women around the globe.

Fine-grained deduplication (also known as delta compression) can achieve a better deduplication ratio compared to chunk-level deduplication. This technique removes not only identical chunks but also reduces redundancies between similar but non-identical chunks. Nevertheless, it introduces considerable I/O overhead in deduplication and restore processes, hindering the performance of these two processes and rendering fine-grained deduplication less popular than chunk-level deduplication to date. In this paper, we explore various issues that lead to additional I/O overhead and tackle them using several techniques. Moreover, we introduce MeGA, which attains fine-grained deduplication/restore speed nearly equivalent to chunk-level deduplication while maintaining the significant deduplication ratio benefit of fine-grained deduplication. Specifically, MeGA employs (1) a backup-workflow-oriented delta selector and cache-centric resemblance detection to mitigate poor spatial/temporal locality in the deduplication process, and (2) a delta-friendly data layout and “Always-Forward-Reference” traversal to address poor spatial/temporal locality in the restore workflow. Evaluations on four datasets show that MeGA achieves a better performance than other fine-grained deduplication approaches. Specifically, MeGA significantly outperforms the traditional greedy approach, providing 10–46 times better backup speed and 30–105 times more efficient restore speed, all while preserving a high deduplication ratio.

This chapter explores the revolutionary application of quantum computing in predicting extreme weather events, a critical aspect of smart disaster management in the era of Industry 6.0. As climate change intensifies the frequency and severity of extreme weather, traditional forecasting methods face limitations in accuracy and computational efficiency. Quantum computing, with its ability to process vast amounts of data and perform complex calculations exponentially faster than classical computers, offers a promising solution. This chapter delves into the theoretical foundations, methodologies, and practical applications of quantum-based predictive modeling for extreme weather events. Through case studies, experimental research, and an analysis of current trends, we demonstrate how quantum algorithms can significantly enhance our ability to forecast and prepare for severe weather phenomena. The potential impact on sustainability and disaster resilience is explored, along with the challenges and future directions in this rapidly evolving field.

Sustainable manufacturing has become a critical priority for organizations seeking to minimize their environmental impact, improve resource efficiency, and enhance their competitiveness. Within this framework, soft computing techniques like swarm intelligence, fuzzy logic, artificial neural networks, and evolutionary algorithms have become effective instruments for managing the uncertainty and complexity present in sustainable manufacturing processes. This chapter offers a thorough exploration of the uses of soft computing in sustainable manufacturing. It begins by introducing the concept of sustainable manufacturing, highlighting its importance, and discussing the key challenges and opportunities. The chapter then presents a summary of various methods for soft computing and their advantages in the sustainable manufacturing domain. The core of the chapter delves into the applications of soft computing in specific areas of sustainable manufacturing, including energy management and optimization, waste management and recycling, supply chain optimization, product design and development, and process monitoring and control. Detailed case studies are provided to illustrate the successful implementation of these techniques in real-world scenarios. Furthermore, the chapter examines the limitations and challenges associated with the integration of soft computing in sustainable manufacturing, such as data availability, system integration, and model interpretability. It also explores the emerging trends and future research directions, emphasizing the potential of hybrid approaches, the integration with Industry 4.0 technologies, the applications of deep learning, and the importance of multidisciplinary collaboration. The findings presented in this chapter underscore the significant potential of soft computing techniques in addressing the sustainability challenges faced by the manufacturing industry. For scholars, professionals, and decision-makers who want to use soft computing to advance the industrial industry’s transition to a more sustainable future, this chapter is an invaluable resource.

The integration of artificial intelligence (AI) and soft computing techniques across various domains has heralded a new era characterized by efficiency, automation, and augmented decision-making capabilities. However, this surge in technological prowess brings with it a commensurate responsibility to address the ethical challenges posed by these transformative innovations. This chapter explores the ethical foundations and frameworks that inform our moral reasoning, delving into consequentialist, deontological, and virtue ethics perspectives. It examines critical ethical issues such as privacy, algorithmic bias, transparency, and human agency in the context of AI and soft computing. Additionally, the chapter explores emerging trends and opportunities, including AI for social good, responsible AI governance, human-AI collaboration, explainable AI, and the role of AI in sustainable development. Furthermore, it discusses the challenges and future directions in harmonizing ethical frameworks, embedding ethics into the AI development lifecycle, addressing ethical challenges in emerging applications, and fostering public trust and acceptance. Ultimately, the chapter underscores the importance of human-centric AI and the need for a collaborative, multidisciplinary approach to navigate the ethical complexities of these transformative technologies.

The real-time monitoring of electromigration in ball-grid-array solder joints is limited to measuring the electrical resistance increase of the solder joints. Tracking the electromigration induced microstructural changes in solder balls requires cross sectioning which is a destructive technique. A novel planar solder geometry was invented and described here that allows real-time, non-destructive monitoring of microstructural changes and the rate of elemental segregation at the anode while simultaneously tracking the extent of electromigration by electrical resistance means. Electromigration in planar geometry tin-bismuth eutectic solder was studied by two means, (a) by the rate of Bi segregation at the anode and (b) by the rate of increase of electrical resistance of the solder, as a function of joint length, solder temperature and electrical current density. At low temperature and low electrical current density there was an extended initial period during which the joint resistance decreased before it increased. At higher temperatures and electrical current densities this initial period of decreasing resistance became less pronounced and at much higher temperature and current density stressing it became non-existent. The rate of bismuth segregation at the anode was somewhat proportional to the solder joint length indicating a probable Blech back-stress effect. Electromigration results from the rate of Bi segregation and the rate of increase of solder joint resistance were summarized using Arrhenius plots. The two plots gave similar electromigration activation energies of 0.7 eV from the electrical measurements and 0.75 eV from the Bi segregation measurements. The Arrhenius plot based on resistance rate increase was also used to predict the electromigration life of Sn-Bi solder joints under typical application conditions.

In each of the last three decades of the 20th century there were unprecedented expansions of sea‐ice over the Labrador Sea basin and influxes of cold fresh water into the subpolar gyre (SPG) which have been described as the Great Salinity Anomalies (GSAs). Employing data for sea surface temperature, salinity, and sea‐ice cover, we propose that these events were downstream consequences of the expansion and subsequent melting of so‐called “Odden” ice formed over the deep basin of the Greenland‐Iceland‐Norway (GIN) Sea in the 1960s, 1970s, and 1980s and additional to the normal East Greenland shelf sea‐ice. We extend previous findings that Odden ice expansions were linked to winter episodes of high atmospheric pressure north of Greenland that directed freezing Arctic winds across the GIN Sea and may also have been associated with increased Arctic sea‐ice volume leading to enhanced ice export through Fram Strait. We show that cold water and ice derived from Odden melting in the summer passed through Denmark Strait and along the East Greenland shelf, and accumulated in the Labrador Sea, creating favorable conditions for winter ice formation during particularly cold years in southwest Greenland. Meltwater from Odden and Labrador Sea ice appeared to break out into the SPG in the fall of 1982 and 1984 respectively and this cold water represents the likely source of the 1982–1985 GSA. These findings further our understanding of the physical processes linking ice formation and melt with ocean circulation in this key component of the climate system.

The fifth generation of mobile networks (5G) brings an evolution of network service provisioning through a new communication paradigm, which enables the development of new applications and improves users’ experience. With 5G, it is envisioned that networks will provide services accessed by a variety of mobile users, some of such services requiring ultra-low latency and very high reliability. 5G also enables massive connectivity of sensors and actuators in the Internet of Things and Cyber-physical Systems scenarios, standing for massive Machine-Type Communication applications. Network Function Virtualization (NFV) is an emerging and promising solution to deal with such demands for flexible and agile service provisioning by replacing dedicated hardware implementations with software instances based on virtualization. A well-known challenge in NFV is the resource allocation problem. In particular, the Service Function Chain (SFC) placement problem is considered a major challenge, even more, when considering the distributed placement in a multi-administrative domain context. We propose a novel solution for the SFC placement problem called Multi-Domain Distributed Auction-Based SFC Placement Algorithm, which relies on an auction-based strategy. The proposed algorithm targets different 5G application scenarios with a focus on solving the SFC placement problem taking into consideration the dynamic aspects of unpredictable requests and the management of the entire auction process to decide which domain will be selected to meet the required SFC. Evaluations were carried out in a simulated environment aiming at assessing the performance of the algorithm in terms of the profit for each service provider during the allocation of the requested services in the respective domain that the service provider is responsible for. Results showed that adopting an auction-based mechanism, and thus allowing a domain to outsource the provision of requested services, successfully reduced the total cost of the service execution in a multi-domain environment. The auction-based approach increases the service provider profit by around 20% in the tested scenarios. Moreover, the number of services placed increases in comparison to the approach where all the services must be executed in the decision domain (with no auction).

Education being an indispensable to our lives, efforts are continuously being made to strengthen it. Assessment lies at the core of any given educational program or module, irrespective of modes (online, offline, blended) and format (short time, part time, or full time). Among the various types of assessment, continuous assessment has been widely adopted. Quizzes are one of the important and common ways to conduct continuous assessment. However, due to the existing processes and involved centralized technologies, there are numerous challenges such as non-transparent and insecure conduction of assessment, non-verifiability of assessment process and computation of final grades from these assessments, and trusted sharing of assessment (or learning) logs along with final grades. Blockchain and smart contracts have emerged as potential technologies which can be leveraged to overcome these challenges. Thus, this paper proposes a blockchain-based architecture for conduction of online quiz. In particular, the paper aims to design and develop secure, transparent, and verifiable, Blockchain and Flutter-based Quiz Mobile-Decentralized Application (BFQM-DApp). To the best of our knowledge, this is first work which attempts to develop a blockchain and flutter-based mobile DApp for online quiz.

With the advent of Industry 4.0, future workspaces are expected to evolve in tandem with technological advances in industry and education. Industry 4.0 calls for transformation and effective talent development is vital in ensuring national aspirations are achieved while eliminating redundancy and ensuring consistency. As such, this study aims to understand the impact of Industry 4.0 on computer engineering-related workforce and skills development within Multinational Companies (MNCs) to Small Medium Enterprises (SMEs) in Malaysia. In this study, online questionnaires were distributed to evaluate the current and future hiring trends. The study reveals that most of the employees have positive perceptions about the industrial current practice on Industry 4.0 and identified the prospective demands on the professions that will be affected. Five significant areas of required competencies found in this study are adaptability, soft skills, software engineering, data analytics, and technical skills. The findings provide empirical evidence about current and future employment scenarios in Malaysia concerning the possible impact of Industry 4.0 on the companies and issues involved in managing the transition to Industry 4.0. Besides, the emergent skills required by workforces that are previously unaddressed in the literature were identified. Empirical evidence from the analysis contributes to shaping the educational systems of the future and helps to proactively identify specific skills shortages at an early stage.

Resilience of 1300nm In(Ga)As/GaAs quantum dot lasers to external optical feedback is systematically investigated from − 10°C to 85 °C. The high-resolution spectral and current-dependent linewidth enhancement factor is measured for all modes in the positive net modal gain region. Complimentary to that, analysis of the free-running and under deployment specific feedback relative intensity noise is discussed in the context of compliance against the IEEE 802.3ah specifications at the full range of temperatures.