Journal of Information, Technology and Policy

Comprehensive Survey of Model Compression and Speed up for Vision Transformers

Feiyang Chen, Ziqian Luo, Lisang Zhou, Xueting Pan, Ying Jiang — Thu, 04 Apr 2024 00:00:00 +0800

Vision Transformers (ViT) have marked a paradigm shift in computer vision, outperforming state-of-the-art models across diverse tasks. However, their practical deployment is hampered by high computational and memory demands. This study addresses the challenge by evaluating four primary model compression techniques: quantization, low-rank approximation, knowledge distillation, and pruning. We methodically analyze and compare the efficacy of these techniques and their combinations in optimizing ViTs for resource-constrained environments. Our comprehensive experimental evaluation demonstrates that these methods facilitate a balanced compromise between model accuracy and computational efficiency, paving the way for wider application in edge computing devices.

Path Selection of Integrated Development of Enterprise Supply Chain Management Driven by Digital Transformation

Xin Yu — Thu, 17 Oct 2024 00:00:00 +0800

Digital transformation provides a new impetus for the development of the supply chain, which can not only improve and expand the visualization level of the supply chain system and the matching scheduling ability, so as to improve the system agility and work efficiency. Based on this, this paper analyzes and discusses the integrated development of supply chain management driven by digital transformation, discusses how to integrate different technologies under the digital background to form a centralized and fast response of supply chain management, and correspondingly cultivate a dynamic supply chain ecosystem to achieve the overall strategic change of the organization.

Influence of E-Waste Management in Green-Computing

Shaheen Manna, Sayantika Mukherjee, Dipanwita Das, Amrita Saha — Tue, 04 Jun 2024 00:00:00 +0800

Over the past two decades, extensive use of electronic devices and rapid urbanization have produced significant electronic waste that pollutes soil, water, and the environment. As a result, environmental activists and scientists around the world now prioritize pollution control and environmental safety above all. One of the by-products of urbanization, electronic waste disposal, has emerged as a foremost social issue. According to the Global E-Waste Monitor 2020, consumers worldwide disposed of 53.6 million tonnes worth of electronics in 2019, an increase of 20% over the previous five years. The gradual deposition of these electronic wastes results in the accumulation of different toxic and heavy metals like lead (Pb) and cadmium (Cd), among others, because these wastes are not biodegradable and contaminate groundwater and soil. In turn, groundwater contamination has an impact on animals, plants, and the entire living system, posing serious health risks and problems. As a result, proper disposal of these electronic wastes has emerged as an urgent need. The process of designing, manufacturing, utilizing, and managing products in an environmentally responsible manner is referred to as "green computing. "E-waste has arisen as a growing environmental issue. The environment and ecology face a problem that cannot be avoided: the use of e-waste. This paper aims to describe e-waste management to implement green computing. The reader of this paper is provided with information about e-waste management and green computing, as well as their potential interactions during the activation process. As a result, e-waste management is working as a green computing strategy.

The Impact of Explainable AI on Customer Trust and Satisfaction in Banking

Yang Ni — Fri, 19 Apr 2024 00:00:00 +0800

This study employed a structured questionnaire to gather data from bank customers, focusing on customer perceptions of Explainable AI, Customer Trust (CT), and Customer Satisfaction (CS) in the banking sector. A total of 180 questionnaires were distributed, with 169 valid responses analyzed. The study selected various indicators such as Customer Age, Education Level, Risk Appetite, Previous Internet Experience, Previous AI Experience, Engagement, Post-Use Feedback, Personalized Service, Prediction Accuracy, Data Privacy, System Reliability, Service Efficiency, and Information Push to assess their impact on customer trust and satisfaction. Reliability and validity analyses ensured the robustness of the collected data. Ordered prohbit analysis revealed significant influences of variables like Customer Risk Preference, Perceived Innovation, and Perceived Accuracy on customer trust and satisfaction.

End-to-End Learning-Based Study on the Mamba-ECANet Model for Data Security Intrusion Detection

Huitao Zhang, Diwei Zhu, Yunxiang Gan, Shuguang Xiong — Sat, 07 Sep 2024 00:00:00 +0800

With the rapid development of information technology, network security issues have become increasingly prominent. In particular, data security intrusions pose serious threats to the data privacy and system security of enterprises and individuals. Traditional intrusion detection systems often exhibit low detection accuracy and high false alarm rates when faced with complex and dynamic network environments and diverse attack methods. Therefore, this paper proposes a data security intrusion detection system based on deep learning, which integrates the Mamba model and ECANet model and employs an end-to-end learning approach for training and optimization. First, the Mamba model is introduced for preliminary data feature extraction, whose efficient feature representation capabilities provide a solid foundation for the subsequent detection process. Then, by integrating the ECANet model, feature selection is further optimized through the attention mechanism, enhancing the model’s focus on important features. Finally, an end-to-end learning approach is adopted to train and optimize the entire system, ensuring excellent performance and robustness in practical applications. Experimental results show that the proposed intrusion detection system demonstrates higher detection accuracy on multiple test datasets, improving by approximately 5% compared to traditional methods, providing a new and effective solution for data security.

Star Map Recognition and Matching Based on Deep Triangle Model

Meng Wang, Huitao Zhang, Ning Zhou — Tue, 14 May 2024 00:00:00 +0800

The star sensor is the key component of Celestial Navigation. It measures the autonomous attitude of navigation bodies by observing stars. And it conducts image collection, preprocessing, feature extraction and matching recognition. Aimed to implement the latter two procedures, we first estimate the coordinate of the point which is the intersection point of the optical axis and the celestial sphere. We employ geometrical knowledge to get the relationship between the intersection point and given projection distances. When distances are unknown, we use Newton’s method to approach the exact coordinate of the intersection point. Based on our coordinate calculation method, we are required to find a principle for improving the accuracy of the coordinate. We first establish a projection screening model to obtain star maps. Then we establish four coordinate systems, i.e., the celestial coordinate system, the star sensor coordinate system, the image coordinate system and the pixel coordinate system. Taking the star map at the north celestial pole as an instance, we finish the transformation of coordinate between different systems and search for the factors affecting accuracy of coordinate. Ultimately, we draw the conclusion that the coordinate accuracy improves, when selected stars projection close the centroid of the photosensitive surface. Aimed to implement the matching recognition, we establish a novel feature extraction and matching model. We take the angle between stars and three of their nearest stars as the feature of the central star. Then we extract the feature matrix of the given star table as the feature database. Using the same way, we get the feature matrix of four–star maps. To achieve the last step of matching recognition, we compare the feature matrix of star maps with the given navigation stars. During the process, we employ DBScan clustering algorithm to implement the matching recognition process. We select the cluster center that satisfies the maximum number of matches as the actual location of the identified star map.