110023 research outputs found
Sort by
Exact assessment of the super Pk-connectivity for the crossed cube interconnection network
No full text[[abstract]]A network is connected if there exists a path between any two distinct vertices. The vertex-connectivity of any connected network is the cardinality of its minimum vertex-cut. Then, a network is super connected if every of its minimum vertex-cuts always consists of a certain vertex’s neighborhood. Kung and Lin (Discret Appl Math 293: 143–156, 2021) recently defined the notion of the super cluster-connectivity as a novel, generalized measure to quantify a network’s connectedness level. This article is dedicated to establishing a deep analysis on the exact formula of super path-connectivity for the crossed cube interconnection network. Accordingly, a sufficient and necessary condition is presented to classify whether or not crossed cubes can be super path-connected
Fuzzy Q-learning Control for Temperature Systems
No full text[[abstract]]In this paper, the reinforcement learning algorithm applied to temperature control of the internet of things (IoT), which aims to develop a multi-purpose intelligent micro-power control switch to achieve advanced temperature control research. This paper is based on the fuzzy Q-learning PID control algorithm based on reinforcement learning, with LinkIt Smart 7688 Duo platform. The error value between the set temperature and the actual sensed temperature is exposed to the reinforcement learning PID control operation. Specifically, a temperature sensor will provide temperature feedback to the LinkIt Smart 7688 Duo in order to achieve the stated temperature control. Finally, the suggested control approach will be compared to PID control to illustrate its efficacy and performance
Emotion-Cause Pair Extraction Based on Machine Reading Comprehension Model
No full text[[abstract]]In this paper, we propose a BERT-based framework for Emotion-Cause Pair Extraction (ECPE) task. Given a passage, the ECPE task aims to jointly extract (1) emotion-related clauses and (2) cause clauses (the clause caused the emotion). Our framework is featured by the following two novel designs. First, we formulate the emotion and cause extraction task as a machine reading comprehension (MRC) task. The MRC task is to read a given text passage, and then answer questions by comprehending the article. In our formulation, we treat the ECPE passage as MRC input and pose questions like (Which clauses cause the emotions?). The idea is to leverage the power of MRC model based on recent pre-trained language model. Second, we formulate the emotion-cause pair detection as contextual relatedness detection problem, which can be also effectively addressed by pre-trained language model. The experiment results based on benchmarking datasets demonstrate the effectiveness of the proposed approach; we advance the state-of-the-art results from 61% to 65% in terms of F1 scores
Real-Time Traffic Speed Estimation for Smart Cities with Spatial Temporal Data: A Gated Graph Attention Network Approach
No full text[[abstract]]Moving vehicles interact with IoT devices deployed in cities and establish social relationships to provide proactive and intelligent services for smart cities. For example, big-data-driven accurate and timely traffic speed prediction systems play an important role in empowering Intelligent Transportation Systems (ITS) in smart cities. The reason is that it is the foundation of modern traffic management and traffic control. Most of the existing advanced traffic speed prediction models are Spatial-Temporal hybrid models. They improve the predicting accuracy by leveraging Graph Convolutional Network (GCN) and Recurrent Neural Network (RNN) to extract spatial and temporal features from the traffic speed data, respectively. However, these models have complex structures and high computational costs. To improve the accuracy of prediction and reduce the cost of model training, we propose a hybrid model, Spatial-Temporal Gated Graph Attention network (ST-GGAN), based on Graph Attention mechanism (GAT) and Gated Recurrent Unit (GRU). Such a method has a simpler structure, lower computational costs, and higher predicting accuracy. The experimental results show that our model's performance is better than the existing advanced models on a real-world dataset
Review on Identity-based Batch Verification Schemes for Security and Privacy in VANETs
No full text[[abstract]]To enhance the quality of vehicular communication, vehicular ad hoc network has to be improved to handle the traffic related issues and maintain privacy. In order to fulfill the same, many schemes have been proposed in last decade. The Identity based Batch verification(IBV) scheme is one such scheme, which makes VANET more secure and efficient. Maintaining privacy through anonymity and reduction of verification time of messages by verifying them in Batch, are the main objectives of this scheme. This paper highlights the security issues of the current IBV scheme and introduces the concept of the random change of Anonymous Identity with time as well as location, to prevent the security attack and to maintain the privacy. In this scheme, performances are evaluated in terms of delay and transmission overhead
Securing heterogeneous embedded devices against XSS attack in intelligent IoT system
No full text[[abstract]]Today, we are living in the realm of Internet of Things (IoT) where simple objects are embedded with the capabilities to understand and operate in its surroundings for offering distinct services to the users. These objects are shipped with their user interfaces that facilitate user to perform administrative activities on the devices using a web browser linked to the device's server. Cross-Site Scripting (XSS) is the most prevalent web application's vulnerability, exploited by an attacker to compromise the embedded devices. This research work is focused towards the development of an approach to defend against XSS attack to safeguard embedded devices deployed in intelligent IoT system. It performs identification through comparing injected strings with the blacklisted attack vectors and mitigates its harmful effects by implementing filtering method in an optimized fashion. It is a fog-enabled approach that operates locally to identify the compromised device within the IoT network. We demonstrate attack exploitation on two smart devices including digital IP Camera and wireless router and then tested the performance of our proposed approach on them. The experimental results highlight the efficacy of the approach as it attains an accuracy of 0.9 and above, on both the tested platforms
Synergic Deep Learning for Smart Health Diagnosis of COVID-19 for Connected Living and Smart Cities
No full text[[abstract]]COVID-19 pandemic has led to a significant loss of global deaths, economical status, and so on. To prevent and control COVID-19, a range of smart, complex, spatially heterogeneous, control solutions, and strategies have been conducted. Earlier classification of 2019 novel coronavirus disease (COVID-19) is needed to cure and control the disease. It results in a requirement of secondary diagnosis models, since no precise automated toolkits exist. The latest finding attained using radiological imaging techniques highlighted that the images hold noticeable details regarding the COVID-19 virus. The application of recent artificial intelligence (AI) and deep learning (DL) approaches integrated to radiological images finds useful to accurately detect the disease. This article introduces a new synergic deep learning (SDL)-based smart health diagnosis of COVID-19 using Chest X-Ray Images. The SDL makes use of dual deep convolutional neural networks (DCNNs) and involves a mutual learning process from one another. Particularly, the representation of images learned by both DCNNs is provided as the input of a synergic network, which has a fully connected structure and predicts whether the pair of input images come under the identical class. Besides, the proposed SDL model involves a fuzzy bilateral filtering (FBF) model to pre-process the input image. The integration of FBL and SDL resulted in the effective classification of COVID-19. To investigate the classifier outcome of the SDL model, a detailed set of simulations takes place and ensures the effective performance of the FBF-SDL model over the compared methods
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
Week-Wise Student Performance Early Prediction in Virtual Learning Environment Using a Deep Explainable Artificial Intelligence
No full text[[abstract]]Early prediction of students’ learning performance and analysis of student behavior in a virtual learning environment (VLE) are crucial to minimize the high failure rate in online courses during the COVID-19 pandemic. Nevertheless, traditional machine learning models fail to predict student performance in the early weeks due to the lack of students’ activities’ data in a week-wise timely manner (i.e., spatiotemporal feature issues). Furthermore, the imbalanced data distribution in the VLE impacts the prediction model performance. Thus, there are severe challenges in handling spatiotemporal features, imbalanced data sets, and a lack of explainability for enhancing the confidence of the prediction system. Therefore, an intelligent framework for explainable student performance prediction (ESPP) is proposed in this study in order to provide the interpretability of the prediction results. First, this framework utilized a time-series weekly student activity data set and dealt with the VLE imbalanced data distribution using a hybrid data sampling method. Then, a combination of convolutional neural network (CNN) and long short-term memory (LSTM) was employed to extract the spatiotemporal features and develop the early prediction deep learning (DL) model. Finally, the DL model was explained by visualizing and analyzing typical predictions, students’ activities’ maps, and feature importance. The numerical results of cross-validation showed that the proposed new DL model (i.e., the combined CNN-LSTM and ConvLSTM), in the early prediction cases, performed better than the baseline models of LSTM, support vector machine (SVM), and logistic regression (LR) models
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