110023 research outputs found
Sort by
PCNNCEC: Efficient and Privacy-Preserving Convolutional Neural Network Inference Based on Cloud-Edge-Client Collaboration
No full text[[abstract]]Deploying convolutional neural network (CNN) inference on resource-constrained devices remains a remarkable challenge for industrial Internet of Things (IIoT). Although the cloud computing shows great promise in machine learning training and prediction, outsourcing data to remote cloud always incurs privacy risk and high latency. Therefore, we design a new framework for efficient and privacy-preserving CNN inference based on cloud-edge-client collaboration (namedPCNNCEC). In PCNNCEC, the model of cloud and the data of client in IIoT are split into two shares and sent to two non-colluded edge servers. By applying the arithmetic secret sharing and pre-computation of beaver's triplets, the two edge servers can jointly calculate the predicting results without learning anything about the model and data. To speed up the pre-computation of offline phase and not sacrifice security, the task of triplets generation is delegated to the cloud, so that the edge servers do not require frequent interactions to generate triplets themselves or introducing additional trusted party. The experimental results show the proposed private comparison protocol achieves a better tradeoff between low latency and high throughput, when it is compared with garbled circuit based protocols and other secret sharing based protocols. Additionally, the benchmarks conducted on realistic MNIST and CIFAR-10 datasets demonstrate that PCNNCEC costs less communication and runtime than two recently related schemes under the same security level
RKD-VNE: Virtual network embedding algorithm assisted by resource knowledge description and deep reinforcement learning in IIoT scenario
No full text[[abstract]]In the era of Industry 4.0, the Industrial Internet of Things (IIoT) is developing rapidly, various IIoT applications pose new challenges to the existing network architecture. On the one hand, these applications put forward higher requirements for the efficient use of network resources. On the other hand, these applications generate massive amounts of information, and they pursue a more secure network environment. Therefore, in order to ensure security while effectively allocating network resources, this paper puts forward a virtual network embedding (VNE) algorithm assisted by resource knowledge description (RKD) and deep reinforcement learning (DRL). First, we use social attribute perception to measure the security of each physical node and regard it as one of the attributes of the physical node. Then, RKD is used to standardize resource constraints before the virtual network is embedded. Finally, the DRL agent derives the probability of the physical node being embedded according to the physical network attributes, and embeds the virtual node according to the probability. Simulation experiments show that compared with the BaseLine algorithm, the RKD-VNE algorithm proposed in this paper has obvious advantages in the general performance of VNE, especially in terms of long-term revenue consumption rate increased by 24.3%
Transfer Learning-Based Multi-Scale Denoising Convolutional Neural Network for Prostate Cancer Detection
No full text[[abstract]]Background: Prostate cancer is the 4th most common type of cancer. To reduce the workload of medical personnel in the medical diagnosis of prostate cancer and increase the diagnostic accuracy in noisy images, a deep learning model is desired for prostate cancer detection.
Methods: A multi-scale denoising convolutional neural network (MSDCNN) model was designed for prostate cancer detection (PCD) that is capable of noise suppression in images. The model was further optimized by transfer learning, which contributes domain knowledge from the same domain (prostate cancer data) but heterogeneous datasets. Particularly, Gaussian noise was introduced in the source datasets before knowledge transfer to the target dataset.
Results: Four benchmark datasets were chosen as representative prostate cancer datasets. Ablation study and performance comparison between the proposed work and existing works were performed. Our model improved the accuracy by more than 10% compared with the existing works. Ablation studies also showed average improvements in accuracy using denoising, multi-scale scheme, and transfer learning, by 2.80%, 3.30%, and 3.13%, respectively.
Conclusions: The performance evaluation and comparison of the proposed model confirm the importance and benefits of image noise suppression and transfer of knowledge from heterogeneous datasets of the same domain
Tree-based Filtering in Pulse-Line Intersection Method Outputs for An Outlier-tolerant Data Processing
No full text[[abstract]]Pulse palpation is one of the non-invasive patient observations that identify patient conditions based on the shape of the human pulse. The observations have been practiced by Traditional Chinese Medicine (TCM) practitioners since thousands of years ago. The practitioners measure the patient’s arterial pulses in three points of both patient wrists called chun, guan, and chy, then diagnose based on their knowledge and experience. Pulse-Line Intersection (PLI) method extract features of each pulse from the observed pulse wave sequence. PLI is performed by summing the number of intersections between the artificial line and the pulse wave. The method is proven in differentiating between hesitant with moderate pulse waves. As the method implemented in Clinical Decision Support System (CDSS) related to pulse palpation, some outlier data might emerge and affect the measurement result. Thus, outlier filtering is needed to prevent unnecessary prediction processes by machine learning (ML) models inside CDSS. This study proposed an outlier filtering model using a decision tree algorithm. This concept is designed by analyzing pulse features values and the chance of odd values combination. Then inappropriate values are excepted using several rules. Every pulse feature list that did not pass the filtering rule is categorized as outliers and were not included for further process. The proposed model works more efficiently than ML models dealing with outliers since this procedure is unsupervised learning with a small number of parameters. Overall, the proposed filtering method can be used in pulse measurement applications by eliminating outlier data that might decrease the performance of ML mode
XSS Armor: Constructing XSS Defensive Framework for Preserving Big Data Privacy in Internet-of-Things (IoT) Networks
No full text[[abstract]]Big data characterizes superfluity of voluminous data that may be in unstructured, structured and/or semi-structured format. Internet of Things networks emerged as one of the biggest sources of big data, stored at the cloud servers. Thereby, it inflates the threats of security attacks such as cross-site scripting (XSS) for stealing the sensitive information and violates the user’s privacy. It leverages an adversary to intervene into user’s individual space. Thus, in this paper, we propose an XSS defensive framework, named as Big IoT Data XSS Armor, to protect user’s privacy in IoT networks. It is a server-side framework that mitigates nonpersistent and persistent XSS attack. Former attack is detected by measuring the similarities in request and response URL with the XSS attack strings. To identify persistent XSS attack, the proposed framework operates by unveiling the irregularities between the genuine features and generated features. The experimental outcomes yield that this framework performs efficiently in shielding against XSS attack and surpasses the detection rate of other existing XSS thwarting techniques because it attains an accuracy of 98.9% and F-measure of 98.4%
台灣證券交易所上市公司月營收之預測-以元大台灣 50 ETF 成分股為例
No full text[[abstract]]股票上市公司的基本面是影響股票市場長期發展趨勢的主因，以長期投資股票市場的角度，上市公司的獲利與盈餘終將反應其市場價值，而營業收入以下簡稱營收正是公司獲利的主要來源。營收具有資訊內涵，因此，上市公司未來月營收的預測，能給予投資者提早取得有意義的資訊內涵。本研究為了進一步提升預測的準確性，對傳統ARIMA模型進行改良，並實施幾種不同演算法的預測效能比較，以平均絕對百分比誤差(MAPE)評估各方法之預測準確度。最後以最佳預測方法預測上市公司未來12個月之月營收，提供股票市場長期投資者事先了解該公司未來月營收可能的成長趨勢及其投資上的參考。[[abstract]]The fundamentals of listed stock companies are the main factors affecting the long-term development trend of the stock market. From the perspective of long-term investment in the stock market, the profits and earnings of listed companies will eventually reflect their market value. Operating revenue has information connotation and isthe main source of company profits. Therefore, the forecast of future monthly revenue of listed companies can give investors early access to meaningful information. In order to further improve the accuracy of forecasting, this study improves the traditional ARIMA model and implements the comparison of the prediction performance of several different algorithms. The average absolute percentage error (MAPE) is used to evaluate the forecast accuracy of each algorithm. Finally, the best algorithms are used to forecast the monthly revenue of the listed company in Taiwan in the next 12 months. The results provide long-term investors in the stock market with a prior understanding of the company's future monthly revenue growth trend and investment reference
A Study on the Relationships Among Hikers�� Leisure Involvement, Place Attachment and Leisure Satisfaction: Taking Yuanlin Tengshan Trail as an Example
No full textA Study on User Pleasure Combining Augmented Reality with Traditional Window Lattice
No full text[[abstract]]文化與教育學習方面，AR技術具有創造互動和愉快性的數位學習體驗。由於學習力有賴於使用者在體驗過程中的互動參與，故本研究藉由AR應用，透過互動功能說明來強化傳統文化內容，以達成或促進使用者主動學習。
然而窗的演變隨著中國傳統建築發展，從簡單實用的功能，邁向實用兼藝術之結合，窗欞格的多風格樣式，除讓建築外觀獲得漂亮的裝飾效果外，尤其在傳統文化的傳承及意涵作用。本研究以傳統中國窗欞格為主題，並以傳統中國窗欞格之台中霧峰林家花園宮保園區為例，彙整出十四種傳統中國窗欞格藝術相關知識學習輔助教材，透過中國窗欞格AR擴增實境APP學習系統創作，讓使用者能夠了解霧峰林家花園窗欞格設計中所包含的涵義與形式，啟發一般大眾探索中國傳統建築窗欞格的動機，了解其美學及珍貴的傳統中國窗欞格之文化價值。
本研究目的為，1）彙整窗欞格圖案基本組成元素；2）分析窗欞格造型圖案之意涵；3）探討性別AR窗欞格配置APP的使用者學習成效；4）探討性別對AR窗欞格配置APP之ARCS學習動機與學習成效差異。研究方法流程：首先製作窗欞格之3D模型設計，完成AR窗欞格APP，並進行「使用AR窗欞格APP」之實驗，受測者共為40人，於實驗後進行使用者前後測與ARCS學習動機滿意度調查問卷。
本研究結果顯示：1）不同性別使用AR窗欞格學習知識APP在學習成效上皆有提升，女生組前後測成績皆高於男生組；2）兩組皆有高度學習動機滿意度，其中女生組學習動機滿意度高於男生組，學習成效男生組相較於女生組低；3）不同性別其學習動機並不影響學習成效。
最後，希冀本研究藉由以台中霧峰林家花園宮保園區，其窗欞格為主題，對AR結合傳統文化之相關設計研究有所貢獻以及參考依據。[[abstract]]AR technology has the potential to create interactive and enjoyable digital learning experiences in cultural and educational learning. As the power of learning depends on the interactive participation of users during the experience, this study uses AR applications to reinforce traditional cultural content through interactive functional illustrations in order to achieve or facilitate active learning.
However, the evolution of windows has evolved with the development of traditional Chinese architecture from a simple and practical function of a practical and artistic combination. In this study, the main theme is traditional Chinese window lattices, and the Taichung Wufeng Lin Family Mansion and Garden is used as an example of traditional Chinese Window Lattice. The augmented reality learning system for Chinese Window Lattices was created to enable users to understand the meaning and form of the Wufeng Lin Family Mansion and Garden Window Lattice design, and to inspire the general public to explore the motivation of traditional Chinese architectural Window Lattices and to understand their aesthetics and the cultural value of the precious traditional Chinese Window Lattice.
The objectives of this study were to 1) compile the basic components of the Window Lattice pattern; 2) analyse the meaning of the Window Lattice pattern; 3) investigate the user learning effectiveness of the gender-specific augmented reality Window Lattice configuration app; 4) investigate the gender-specific learning motivation and effectiveness of the ARCS for the augmented reality Window Lattice configuration app. (4) To investigate the difference in motivation and learning effectiveness of gender on the ARCS of the augmented reality Window Lattice configuration app. Methodology: Firstly, a 3D model of Window Lattice was created, and the augmented reality Window Lattice App was completed.
The results of this study showed that: 1) the use of the Expanded Reality Window Lattice Learning Knowledge App improved learning outcomes for both genders, with the girls' group scoring higher than the boys' group on the pre and post tests; 2) both groups had high levels of motivation satisfaction, with the girls' group scoring higher than the boys' group, and learning outcomes were lower in the boys' group compared to the girls' group; and 3) motivation did not affect learning outcomes across genders.
Finally, it is hoped that this study will contribute to and inform research on the integration of traditional culture into augmented reality through the use of the Window Lattice as a theme in the Taichung Wufeng Lin Family Mansion and Garden
A Study on the Essence of Virtual Characters
No full text[[abstract]]角色對故事創作上是相當重要元素之一，即使虛構想像是如超能力般的脫離現實與非科學，其魅力依舊不減吸引著你對其產生共鳴與情感。數位化時代與新媒體（new media）的轉變下，相關「虛擬角色」的現象在生活與文化、經濟等各層面，都連結著「存在」與「相信」的心理認知之意義與價值。因此，本研究是解析虛擬角色的本質為主要目的，以訊息的表達與意義兩主軸出發，導入皮爾斯（Santiago Peirce）的三位一體（triaque）符號理論為基礎。文獻探討從認識論、方法論、再到本體論的科學哲學為研究取徑，探討意義表達的過程與情感層次的階段與轉變；從而歸納出「形式意義」到「修辭意義」再到「象徵意義」三階段之「虛擬角色符號論」。透過「混和研究」法整合量化與質性的調查，量化資料分別以統計分析，質性資料以開放編碼，抽取「共同構念」繪製共識地圖（consensus map）；再相互交叉驗證得出以下結論：虛擬角色的本質是一種「虛擬人格化意識狀態」，當中包含了：「訊息的人格化」、「人格化意識化」、「人格的意識精神化」三個層次。而認知過程是符合「虛擬角色符號論」的三個階段的意義，本質上分別具有以下三個特性：1. 形式意義是「人格化訊息的人工文化符號創作」；2. 修辭意義是「角色訊息的意義交換與形式意義的認知分離，最終形成概念性的人格化意識」；3. 象徵意義是「群體性人格化意識的精神昇華，同時成為文化代表性的精神象徵」；並以「符號認知」與「語境溝通」之思考，提出的虛擬角色設計法則作為本研究之貢獻，提供相關領域的創作與研究之參考。[[abstract]]Characters are critical elements in story creation. Although most characters are conceived through the imagination and are, therefore, fictional— with superpowers not based on science—, they still resonate with us and speak to our emotions. The new media in the digital age has influenced the development of "virtual characters", their cultural and economic value is fundamentally a cognitive process of "existence" and "belief", which is related to the essence of cognitive meaning. Therefore, the aim of this study was to analyze the essence and cognitive psychology of virtual characters. Based on the Charles Sanders Peirce's theory of semiotics, the focus of the study was message and signification. The literature review was centered on the process of signification and the transition and phases of emotional layers through epistemology, methodology, and ontology in the philosophy of science, concluding with three phases of the “semiotics of virtual characters,” including “formal meaning,” “rhetorical meaning,” and “symbolic meaning.” Through mixed methods research, quantitative and qualitative data were integrated. The quantitative data were analyzed statistically, and common ideas were extracted from the qualitative data through open coding to create a consensus map. After cross-validation of the results, the following conclusion was reached. The essence of virtual characters is the state of a “virtual personified consciousness,” which includes “personification of the message,” “personification of consciousness,” and the “spiritualization of consciousness.” The cognitive process corresponds to the meaning of the three stages of the semiotics of virtual (artificial) cultural sign creation of personified messages characters with the following characteristics: (1) the formal meaning is “the,” (2) the rhetorical meaning is “the cognitive dissociation between meaning exchanges of characters’ messages and formal meaning; a conceptual personified consciousness would be formed eventually,” and (3) the symbolic meaning is “the spiritual sublimation of collective personified consciousness; meanwhile becoming a symbol of cultural spirit.” Based on these characteristics, principles and methods of creating virtual characters are proposed as references for creative work and research in relevant fields
Construction and Research of Web Application in Small Catering Ordering System
No full text[[abstract]]由近年來的新冠疫情已經對中小型餐飲業者造成了嚴重的影響，不僅使人們面臨生命危險，也造成了巨大的經濟損失。由於無法預測疫情會持續多久，因此本研究提出一個系統，在配合防疫的情況下幫助中小型餐飲業者恢復生意。在人們高度聚集的情況下，如排隊、點餐和結帳等情況下，感染的風險更高。因此，在這個特殊時期，我們需要減少不必要的人體接觸。
本研究的主要目標是開發一個能提高中小型餐飲業者的點餐和結帳效率的Web Application小型餐飲點餐系統。這個系統有助於創造一個低接觸的環境，以減少客戶的感染風險，並且系統化並高效的管理日常運作。當客戶在餐廳進行點餐和結帳時，他們不必在人群中逗留太長的時間，這可以有效的降低感染的風險。而餐廳還可以通過這個系統管理訂單和收益，進而提高餐廳的營運效率。
只要連上網，這個系統可以在任何的智能終端上使用，不需進行任何額外的安裝並且可以在手機、平板或電腦上進行點餐和結帳。客戶可直接通過手機或電腦在餐廳的線上菜單選擇他們想要的餐點，下單以及結帳。這個系統不僅可以減少需要排隊的情況，還可以使客戶快速地完成點餐和結帳的流程。這將有助於提高客戶的滿意度，並且確保客戶的安全。另外，餐館可以通過系統管理他們的訂單和庫存，提高營業效率。[[abstract]]Due to the pandemic in the last few years, the food industry has been heavily affected. As for now we can’t be sure how long this pandemic will last, thus we need to come out with a solution to counter this situation. The epidemic not only exposed people to danger but also caused huge economic loss. During this period, people tend to be exposed to greater risks in some situations of high gathering activities such as gathering, doing payment, food ordering, and queuing up. Due to the pandemic being highly contagious, therefore it's safer for people to reduce unnecessary human contact at this moment.
The main goal of this research is to develop a web-based cloud ordering system application platform that can improve the ordering and checkout efficiency of small and medium-sized restaurants. This system helps to create a low-touch environment to reduce the risk of infection for customers and manage daily operations systematically and efficiently. When customers order and checkout at the restaurant, they don't have to stay in the crowd for too long, which can effectively reduce the risk of infection. The restaurant can also manage orders and revenue through this system, thereby improving the operating efficiency of the restaurant.
This system can be used on any smart terminal without any additional installation, and can order and pay on mobile phones, tablets or computers as long as it is connected to the Internet. Customers can choose the meal they want at the restaurant and check out directly through their mobile phone. This system not only reduces the need to queue, but also allows customers to quickly complete the ordering and checkout process. In addition, restaurants can manage their orders and inventory through the system, improving business efficiency