Archives

  • Vol. 1 No. 1 (2023)

    Editorial Introduction to the First Issue of Cambridge Explorations in Arts and Sciences (CEAS)

    On behalf of the Editorial Team, I would like to welcome you to the inaugural issue of Cambridge Explorations in Arts and Sciences (CEAS). As Editor-in-Chief, I am delighted to announce that the first issue is focussed on Artificial Intelligence (AI) in a range of applications and social settings.

    CEASC has a unique mission. We aim to give a platform to High School and University students who would like to gain early experience in publishing their working papers and projects so as to reach a wide audience. The papers are by authors that are all outstanding in their cohort, in terms of content, organisation, and presentation. Showcasing a publication in CEASC will benefit them in many ways. Firstly, they will gain a public platform for their work through which they will be able to put forward their ideas, arguments, designs, and models: these will be on show for all to see. A corresponding benefit is that it will be of great help in their progression through the educational system and the world of work. In today’s fast-moving environment, it is essential to publish as soon as possible. Naturally, this is to claim a stake in the debate and to assert ownership of intellectual property of all types. That is why we offer a fast-publishing model while retaining academic standards. CEASC is intended to be for authors from a broad range of academic disciplines: from business to technology, from history to sociology. The focus of our first issue is AI. However, many papers are not only technical in nature but also examine the social impacts of such technology that are vital for people in general to understand and appreciate.

    Our First Papers

    Our first ten papers are dedicated to the exploration of AI in a broad range of social contexts ranging from healthcare to song generation. They have been written by university students on a Cambridge educational programme (please contact the Editor-in-Chief for details).

    Gao Shuyan et. al. provide an AI method for finding the optimal battery material suitable for an artificial heart, while Yiting Zheng et. al. not only create a predictive model of breast cancer by analysing methylation molecular markers but also examine public attitudes to using AI for the same. Also in the medical sphere, Kexin Yuan et. al. propose a new idea for colorectal cancer (CRC) screening and diagnosis based on three comparative AI analytical approaches of intestinal flora of CRC. Ziyin Wang et al. apply a U-net model to retinal vessel segmentation and achieve data enhancement in morphology methods to highlight the importance of data preprocessing in digital image processing, and Yuyan Zhang, Shuoyue Feng, and Yiming Wang propose a model of ambulance routing that maximises not only speed to hospital but allows for consideration of hospital resources across locations to improve patient outcomes. In the arts, Shuoshuo Yin et al. evaluate AI-generated art in Tiktok and other platforms both from a technical point of view and that of human artists, and Siyuan Chen et.al. investigate the capability of AI to generate songs and lyrics for use in the primary school as part of a STEAM curriculum. Relevant to all, Zixuan Tang et. al. use AI to make correlations between air pollution indices and expressions of happiness on social media with suggestions for a new indicator, while dining experiences might be enhanced by the AI model for the optimisation of cost control regarding key factors for individual buffet restaurants and chains proposed by Shuang Wu et. al. Finally, Yuhang Pan et al. provide a technical account of the Deep Q-Network and how it can be improved to get better outcomes in the Snake game. All in all, we offer a fascinating collection of articles in our first edition and we hope the reader will enjoy them!

     

    Mark Perkins

    Editor-in-Chief

  • Vol. 1 No. 2 (2023)

    Editorial Introduction to the October 2023 Issue of Cambridge Explorations in Arts and Sciences (CEAS)

    On behalf of the Editorial Team, I would like to welcome you to the October issue of Cambridge Explorations in Arts and Sciences (CEAS). As Editor-in-Chief, I am delighted to announce that this issue is wide-ranging with topics from technical fields such as machine learning and artificial intelligence (AI) as applied to viral spread, deep neural networks, financial technology, image reconstruction, driver assistance and parking assistance, robotic arm improvement, disease modelling, and plant conservation. On the social sciences side, topics include digital museums, and the nature-nurture debate. As with our June 2023 issue, the papers have been written by university students on a Cambridge educational programme (please contact the Editor-in-Chief for details).

     

    Our Eleven Papers

    In the first of our eleven papers, Xiao Yuxuan et al. revisit the nature-nurture debate and generate novel results by conducting inter-cultural fieldwork based on a combined online-offline questionnaire.  Then, Wang Mingjie et al. create a dynamic model to analyse the behaviour of an epidemic by linking infecting to infected people with the goal of limiting viral spread, while Xia Mingwei et al. conduct a thorough investigation of users’ experiences of the digital museum as compared with the physical experience. They identify many gaps where the digital experience can be improved. Taking us into nature conservation, Deng Chuyu et al. examine traditional ecosystem mapping methods and compare them to results generated by the CAPTAIN artificial intelligence-based system. They make a strong case that the AI approach supersedes the traditional and represents a leap forward. Next comes medical technology, advancing rapidly. Robotic arms are an important part of neurorehabilitation and Chen Wenyi et al. propose an EEG-based control method to enhance their function. With many applications, the paper by Xu Xiaotang et al. is specifically focussed on deep neural networks and they evaluate single and ensemble attack methods. Xu Yingche et al., on the other hand, tackle the issue of image reconstruction and how it can be enhanced by a combined use of VAEs and Planar Flows drawing on machine learning. Driverless cars are becoming all the more common. Shi Yutian et al. deploy artificial intelligence for assisted driving and put forward certain solutions to specific problems encountered today in the area of safety that are achievable by harmonising existing advanced driver assistance and driver state monitoring systems. Continuing in the theme of driving, Li Wenxuan et al. address the problem faced by all drivers - where to park. They modify a ResNet model with transfer learning to produce a solution that helps drivers in the smart city and potentially eases congestion. Turning to the world of finance, Yuan Yiwei et al. examine the use of the sandbox as a tool for evaluating potential financial regulation. As a result of a comparison of current use in the UK, Singapore, and China they make concrete proposals as to how the Chinese process could be improved. Going further in the area of financial technology, Cheung Arts et al. review the use of cryptocurrencies in Hong Kong and China and make policy recommendations for their future use in China.

    Mark Perkins

    Editor-in-Chief

  • Vol. 2 No. 1 (2024)

     

    Editorial Introduction to the January 2024 Issue of Cambridge Explorations in Arts and Sciences (CEAS) 

    On behalf of the Editorial Team, I would like to welcome you to the January issue of Cambridge Explorations in Arts and Sciences (CEAS), our first of 2024. As Editor-in-Chief, I am pleased to see that this issue has a large spread of topics.  These include the impact of AI on the automotive sector, AI-enhanced walking aids for the elderly, and developments in remote sensing of small images. AI in economic decision-making is also tackled, as is AI and copyright. Materials science is represented through a nanoscale study of the properties of copper, and Bayesian models are examined with potential applications to neural networks. On the social sciences side, topics include a different perspective on digital museums, the nature of student discourse in the dormitory setting, and the use of AI in children’s painting. As with our previous issues, the papers have been written by university students on a Cambridge educational programme (please contact the Editor-in-Chief for details). 

     

    Our Eleven Papers 

    In their study of the automaker Haier, Wu Jingyi et al. found that AI applications have a powerful influence on R&D, productivity, defect rate, and carbon footprint, leading to an overall improvement in efficiency. Also in the automotive sector, Dai Le et al. examined the impact of artificial intelligence on the development of the carmaker BYD Co. AI appeared to improve all indicators, although the authors pointed to the need for a longer study than the 7 years of data available to them could afford. Then, Jiang Yulin et al. evaluated several walking aids for the elderly and suggested not only AI-enhanced improvements but also commercial prototypes. 

    Kong Ziyang et al. tackled the difficult issue of remote sensing of small objects. They reviewed existing technologies and proposed improvements including a new visual interface. Zhu Yukai et al. address the interesting question of whether AI processing of data affects international economic policymaking. They also discuss the risks of AI use in formulating international economic policies from an ethical perspective, a hitherto neglected area. With the use of AI, Ling Hongyi et al. trained a convolutional neural network on copyright data to identify infringement. The model created gave an accuracy of 80%. Tang Minyu et al. made a detailed investigation of the properties of copper at the nanoscale level. The study contributes significantly to the understanding of copper’s mechanical properties and the application of computational simulation in material. In a highly technical paper, Chen Yankun et al. looked at the Auto-Encoding Variational Bayes (AEVB) estimator. Using the standard stochastic gradient method, they found that the estimator can be directly discretized and optimized, improving the existing dynamic Bayesian network and achieving substantial improvements in accuracy and efficiency relevant to time-series modeling and classification. 

    On the humanistic side, Zhang Tianshu et al. present the results of a market analysis of young visitors to museums and show how a digital approach to the visitor experience will not only be beneficial but also possibly lead to increased employment. Next, from a linguistic and sociological point of view, Haoyang Guan et al. conducted a questionnaire of Chinese students in dormitories to investigate the discourse of conflict expressed by gender. They found clear differences as to the perception of conflict and its resolution mechanisms. Finally, Ye Lyuzhaozhao et al. conducted research on the use of AI as an assistant for children when making their paintings. The research was based on one in-depth case study combined with a questionnaire sent to adults. It was found that the child enjoyed using an AI tool and the results were evaluated positively overall by adults leaving scope for future use. 

    I hope the reader will enjoy the diversity of research in our January issue! 

    Mark Perkins 

    Editor-in-Chief