Computer Science Review最新文献

The variety of enormous information on the web encourages the field of recommender systems (RS) to flourish. In recent times, deep learning techniques have significantly impacted information retrieval tasks, including RS. The probabilistic and non-linear views of neural networks emerge to generative models for recommendation tasks. At present, there is an absence of extensive survey on deep generative models for RS. Therefore, this article aims at providing a coherent and comprehensive survey on recent efforts on deep generative models for RS. In particular, we provide an in-depth research effort in devising the taxonomy of deep generative models for RS, along with the summary of state-of-art methods. Lastly, we highlight the potential future prospects based on recent trends and new research avenues in this interesting and developing field. Public code links, papers, and popular datasets covered in this survey are accessible at: https://github.com/creyesp/Awesome-recsys?tab=readme-ov-file#papers.

Natural Language-based Source Code Generation (NLSCG) holds the promise to revolutionize the way how software is developed by means of facilitating a collection of intelligent technical enablers, based on sustained improvements on the natural language to source code pipelines and continuous adoption of new coding paradigms. In recent years, a large variety of NLSCG technical solutions have been proposed, and quite exciting experimental results have been reported. Meanwhile, current researches and initiative application projects in this area reflect a large diversity of NLSCG contexts and of major technical enablers. Such heterogeneity, fragmentation, and vagueness of the NLSCG technical landscape are currently frustrating the full realization of the NLSCG research and application vision. Players in this field could not find systematic guidelines on how to effectively address the ”known unknowns” and how to simply spot the ”unknown unknowns”, which eventually hinder the turning of NLSCG solutions into further research enhancements or production applications. Understanding the context, boundaries, capabilities, and integrations of NLSCG enablers is considered as one of the key drivers for the more practical application of NLSCG models. In this paper, we analyze in detail the natural language to source code pipelines and the evolvement of source code generation tasks, by considering both the problem context and technological aspects. A foresight reference framework for NLSCG is proposed to help handle the source code generation tasks with proper intelligent models. We review the present-day NLSCG technical landscape, as well as the core technical enablers along the source code generation pipelines. Relevant experiments are conducted to validate the role of representative models across different technical enablers on typical datasets, and we finally highlight the contribution of different enablers to code generation capabilities.

Raspberry Pi is an invaluable and popular prototyping tool in scientific research for experimenting with a wide variety of ideas, ranging from simple to complex projects. This review article explores how Raspberry Pi is used in various studies, discussing its pros and cons along with its applications in various domains such as home automation, agriculture, healthcare, industrial control, and advanced research. Our aim is to provide a useful resource for researchers, educators, students, product developers, and enthusiasts, helping them to grasp the current status and discover new research possibilities using Raspberry Pi.

This paper provides an extensive and in-depth survey of behavior modeling for complex intelligent systems, focusing specifically on the innovative applications of Generative Adversarial Networks (GANs). The survey not only delves into the fundamental principles of GANs, but also elucidates their pivotal role in accurately modeling the behaviors exhibited by complex intelligent systems. By categorizing behavior modeling into prediction and learning, this survey meticulously examines the current landscape of research in each domain, shedding light on the latest advancements and methodologies driven by GANs. Furthermore, the paper offers insights into both the theoretical underpinnings and practical implications of GANs in behavior modeling for complex intelligent systems, and proposes potential future research directions to advance the field. Overall, this comprehensive survey serves as a valuable resource for researchers, practitioners, and scholars seeking to deepen their understanding of behavior modeling using GANs and to chart a course for future exploration and innovation in this dynamic field.

The integration of multiple heterogeneous data into graph models has been the subject of extensive research in recent years. Harnessing these resulting Heterogeneous Information Networks (HINs) is a complex task that requires reasoning to perform various prediction tasks.

In the last decade, multiple Artificial Intelligence (AI) approaches have been developed to bridge the gap between the abundance of diverse data within various fields, their heterogeneity and complexity within HINs. A focus has been directed on developing graph-oriented algorithms that can effectively analyze and leverage the rich information in HINs.

Given the sheer volume of approaches being developed, selecting the most suitable one for a specific objective has become a daunting challenge. This article reviews the recent advances in AI methods for modeling and analyzing HINs. It proposes a cartography of these approaches, structured as a pipeline, offering diverse options at each stage. This structured framework aims to guide practitioners in choosing the most fitting methods based on the nature of their data and specific objectives.

Cross-site scripting (XSS) is one of the major threats menacing the privacy of data and the navigation of trusted web applications. Since its disclosure in late 1999 by Microsoft security engineers, several techniques have been developed with the aim of securing web navigation and protecting web applications against XSS attacks. XSS has been and is still in the top 10 list of web vulnerabilities reported by the Open Web Applications Security Project (OWASP). Consequently, handling XSS attacks has become one of the major concerns of several web security communities. Despite the numerous studies that have been conducted to combat XSS attacks, the attacks continue to rise. This motivates the study of how the interest in XSS attacks has evolved over the years, what has already been achieved to prevent these attacks, and what is missing to restrain their prevalence. In this paper, we conduct a systematic mapping and a comprehensive survey with the aim of answering all these questions. We summarize and categorize existing endeavors that aim to handle XSS attacks and develop XSS-free web applications. The systematic mapping yielded 157 high-quality published studies. By thoroughly analyzing those studies, a comprehensive taxonomy is drawn out outlining various techniques used to prevent, detect, protect, and defend against XSS attacks and vulnerabilities. The study of the literature revealed a remarkable interest bias toward basic (84.71%) and JavaScript (81.63%) XSS attacks as well as a dearth of vulnerability repair mechanisms and tools (only 1.48%). Notably, existing vulnerability detection techniques focus solely on single-page detection, overlooking flaws that may span across multiple pages. Furthermore, the study brought to the forefront the limitations and challenges of existing attack detection and defense techniques concerning machine learning and content-security policies. Consequently, we strongly advocate the development of more suitable detection and defense techniques, along with an increased focus on addressing XSS vulnerabilities through effective detection (hybrid solutions) and repair strategies. Additionally, there is a pressing need for more high-quality studies to overcome the limitations of promising approaches such as machine learning and content-security policies while also addressing diverse XSS attacks in different languages. Hopefully, this study can serve as guidance for both the academic and practitioner communities in the development of XSS-free web applications.

This review paper offers an in-depth analysis of AI-powered virtual conversational agents, specifically focusing on OpenAI’s ChatGPT. The main contributions of this paper are threefold: (i) an exhaustive review of prior literature on chatbots, (ii) a background of chatbots including existing chatbots/conversational agents like ChatGPT, and (iii) a UI/UX design analysis of prominent chatbots. Another contribution of this review is the comprehensive exploration of ChatGPT’s applications across a multitude of sectors, including education, business, public health, and more. This review highlights the transformative potential of ChatGPT, despite the challenges it faces such as hallucination, biases in training data, jailbreaks, and anonymous data collection. The review paper then presents a comprehensive survey of prior literature reviews on chatbots, identifying gaps in the prior work and highlighting the need for further research in areas such as chatbot evaluation, user experience, and ethical considerations. The paper also provides a detailed analysis of the UI/UX design of prominent chatbots, including their conversational flow, visual design, and user engagement. The paper also identifies key future research directions, including mitigating language bias, enhancing ethical decision-making capabilities, improving user interaction and personalization, and developing robust governance frameworks. By solving these issues, we can ensure that AI chatbots like ChatGPT are used responsibly and effectively across a broad variety of applications. This review will be a valuable resource for researchers and practitioners in understanding the current state and future potential of AI chatbots like ChatGPT.

Distributed Denial of Service (DDoS) attacks in IoT networks are one of the most devastating and challenging cyber-attacks. The number of IoT users is growing exponentially due to the increase in IoT devices over the past years. Consequently, DDoS attack has become the most prominent attack as vulnerable IoT devices are becoming victims of it. In the literature, numerous techniques have been proposed to detect IoT-based DDoS attacks. However, techniques based on Artificial Intelligence (AI) have proven to be effective in the detection of cyber-attacks in comparison to other alternative techniques. This paper presents a systematic literature review of AI-based tools and techniques used for analysis, classification, and detection of the most threatening, prominent, and dreadful IoT-based DDoS attacks between the years 2019 to 2023. A comparative study of real datasets having IoT traffic features has also been illustrated. The findings of this systematic review provide useful insights into the existing research landscape for designing AI-based models to detect IoT-based DDoS attacks specifically. Additionally, the study sheds light on IoT botnet lifecycle, various botnet families, the taxonomy of IoT-based DDoS attacks, prominent tools used to launch DDoS attack, publicly available IoT datasets, the taxonomy of AI techniques, popular software available for ML/DL modeling, a list of numerous research challenges and future directions that may aid in the development of novel and reliable methods for identifying and categorizing IoT-based DDoS attacks.

Clustering is a fundamental data exploration technique to discover hidden grouping structure of data. With the proliferation of big data, and the increase of volume and variety, the complexity of data multiplicity is increasing as well. Traditional clustering methods can provide only a single clustering result, which restricts data exploration to one single possible partition. In contrast, multiple clustering can simultaneously or sequentially uncover multiple non-redundant and distinct clustering solutions, which can reveal multiple interesting hidden structures of the data from different perspectives. For these reasons, multiple clustering has become a popular and promising field of study. In this survey, we have conducted a systematic review of the existing multiple clustering methods. Specifically, we categorize existing approaches according to four different perspectives (i.e., multiple clustering in the original space, in subspaces and on multi-view data, and multiple co-clustering). We summarize the key ideas underlying the techniques and their objective functions, and discuss the advantages and disadvantages of each. In addition, we built a repository of multiple clustering resources (i.e., benchmark datasets and codes). Finally, we discuss the key open issues for future investigation.

The growth rate in big data and internet of things (IoT) is far exceeding the computer performance rate at which modern processors can compute on the massive amount of data. The cluster and cloud technologies enriched by machine learning applications had significantly helped in performance growths subject to the underlying network performance. Computer systems have been studied for improvement in performance, driven by user’s applications demand, in the past few decades, particularly from 1990 to 2010. By the mid of 2010 to 2023, albeit parallel and distributed computing was omnipresent, but the total performance improvement rate of a single computing core had significantly reduced. Similarly, from 2010 to 2023, our digital world of big data and IoT has considerably increased from 1.2 Zettabytes (i.e., sextillion bytes) to approximately 120 zettabytes. Moreover, in 2022 cloud datacenters consumed $\sim$ 200TWh of energy worldwide. However, due to their ever-increasing energy demand which causes ${\mathrm{CO}}_2$ emissions, over the past years the focus has shifted to the design of architectures, software, and in particular, intelligent algorithms to compute on the data more efficiently and intelligently. The energy consumption problem is even greater for large-scale systems that involve several thousand servers. Combining these fears, cloud service providers are presently facing more challenges than earlier because they fight to keep up with the extraordinary network traffic being produced by the world’s fast-tracked move to online due to global pandemics. In this paper, we deliberate the energy consumption and performance problems of large-scale systems and present several taxonomies of energy and performance aware methodologies. We debate over the energy and performance efficiencies, both, which make this study different from those previously published in the literature. Important research papers have been surveyed to characterise and recognise crucial and outstanding topics for further research. We deliberate numerous state-of-the-art methods and algorithms, stated in the literature, that claim to advance the energy efficiency and performance of large-scale computing systems, and recognise numerous open challenges.