With the development of smart cities and the increase in vehicles, traffic accident detection has become a crucial component of road safety monitoring in smart cities. Existing traffic accident detection methods suffer from an imbalance between speed and accuracy, poor adaptability to diverse scenarios, and insufficient generalization capabilities, making it challenging to meet real-time monitoring requirements. In this study, a YOLOv5-based traffic accident detection method is proposed to enhance system efficiency in complex road environments. The research leverages the deep learning object detection framework YOLOv5 and optimizes model parameters and training strategies to efficiently identify and localize traffic accidents. The experiments conducted on a self-constructed traffic accident dataset demonstrate that the proposed model achieves superior performance in accuracy and speed, with a of 90.93%, while maintaining high inference speed and meeting real-time detection requirements. The study provides an effective solution for traffic accident detection. It validates the feasibility of applying the YOLOv5 model in road scenarios, offering a significant reference for traffic safety monitoring systems in smart cities.

Human Activity Recognition is a promising area having potential to benefit the human society by developing assistive technologies in order to aid elderly, chronically ill and for people with special needs. Accurate activity recognition is challenging because human activity is complex and highly diverse. A comparative study on Human Activity Recognition (HAR) dataset based on four methods, Random Forest (RF), Support Vector Machines (SVM), K-Nearest Neighbors (KNN), and Linear Discriminant Analysis (LDA) with different features for selecting the best classifier among the models to test the dataset, has been carry out in this paper. The best classifier is choosing by accuracy of the model. We compare the result of dataset with and without important features selection by RF methods varImp(), Boruta, and Recursive Feature Elimination (RFE) to get the best accuracy. From the four methods, we found that the method RF have high accuracy from every group (98.16%, 98.09%, 93.6%), which is considered as a best classifier.

Deep learning is a methodology applied across many fields. User comments are important for recommender systems because they include various types of emotional information that may influence the correctness or precision of the recommendation. Improving the accuracy of user ratings from obtained feasible recommendations is essential. In this paper, we propose a deep learning model to process user comments and to generate a possible user rating for user recommendations. First, the system uses sentiment analysis to create a feature vector as the input nodes. Next, the system implements noise reduction in the data set to improve the classification of user ratings. Finally, a deep belief network and sentiment analysis (DBNSA) achieves data learning for the recommendations. The experimental results indicated that this system has better accuracy than traditional methods.

In pattern recognition and data mining, clustering is a classical technique to group matters of interest and has been widely employed to numerous applications. Among various clustering algorithms, K-means (KM) clustering is most popular for its simplicity and efficiency. However, with the rapid development of the social network, high-dimensional data are frequently generated, which poses a considerable challenge to the traditional KM clustering as the curse of dimensionality. In such scenarios, it is difficult to directly cluster such high-dimensional data that always contain redundant features and noises. Although the existing approaches try to solve this problem using joint subspace learning and KM clustering, there are still the following limitations: 1) the discriminative information in low-dimensional subspace is not well captured; 2) the intrinsic geometric information is seldom considered; and 3) the optimizing procedure of a discrete cluster indicator matrix is vulnerable to noises. In this paper, we propose a novel clustering model to cope with the above-mentioned challenges. Within the proposed model, discriminative information is adaptively explored by unifying local adaptive subspace learning and KM clustering. We extend the proposed model using a robust l2,1-norm loss function, where the robust cluster centroid can be calculated in a weighted iterative procedure. We also explore and discuss the relationships between the proposed algorithm and several related studies. Extensive experiments on kinds of benchmark data sets demonstrate the advantage of the proposed model compared with the state-of-the-art clustering approaches.

Cognitive radio (CR) allows unauthorized users to access authorized band without interfering the authorized user, thereby improving the bandwidth efficiency. In addition, cooperative communications with relay stations (RSs) can be used to improve throughput performance of 4G downlink network. In this study a cognitive radio assisted cooperation (CRAC), which combines advantages of CR and cooperative communications with RSs, is considered for resources allocation in the downlink orthogonal frequency division multiple access (OFDMA) networks. Two adaptive resource allocation algorithms are presented and experimental results demonstrate that the proposed algorithms can not only enhance throughput, but also improve fairness and utility of users assuming available channels are known by base station.

In this paper, both effective bandwidth extension and distributed issues are investigated to improve the energy efficiency for the orthogonal frequency division multiplexing (OFDM) wireless communication systems. A Fuzzy based energy efficient bandwidth expansion (FBEEBE) scheme is proposed to fast infer the bandwidth extension factors (BEFs) for users based on fuzzy inference systems. The proposed fuzzy inference system can infer the BEFs from the two inputs of SNR and bandwidth requirement of users. Simulation results show that the proposed FBEEBE can improve the energy efficiency comparing to that of same bandwidth extension (SBE) scheme.

Diabetes mellitus is a common chronic disease in recent years. According to the World Health Organization, the estimated number of diabetic patients will increase 56% in Asia from the year 2010 to 2025, where the number of anti-diabetic drugs that doctors are able to utilize also increase as the development of pharmaceutical drugs. In this paper, we present a recommendation system for anti-diabetic drugs selection based on fuzzy reasoning and ontology techniques, where fuzzy rules are used to represent knowledge to infer the usability of the classes of anti-diabetic drugs based on fuzzy reasoning techniques. We adopt the "Medical Guidelines for Clinical Practice for the Management of Diabetes Mellitus" provided by the American Association of Clinical Endocrinologists to build the ontology knowledge base. The experimental results show that the proposed anti-diabetic drugs recommendation system gets the same accuracy rate as the one of Chen et al.'s method (R. C. Chen, Y. H. Huang, C. T. Bau and S. M. Chen, Expert Syst. Appl.39(4) (2012) 3995–4006.) and it is better than Chen et al.'s method (R. C. Chen, Y. H. Huang, C. T. Bau and S. M. Chen, Expert Syst. Appl.39(4) (2012) 3995–4006.) due to the fact that it can deal with the semantic degrees of patients' tests and can provide different recommend levels of anti-diabetic drugs. It provides us with a useful way for anti-diabetic drugs selection based on fuzzy reasoning and ontology techniques.

In the recent year, the position location system with ubiquitous computing has become very important, and the use of technology in the position location system has increasingly become the object of study and enterprise applications. One of the rapidly advancing technologies of position location system research is the global positioning system (GPS) but in indoor environments, the receiver may not receive the signal because the signal is subject to the building's impact. This congenital limitation renders the GPS unusable for the indoor position location system. In this paper, we will use cascade correlation network for an indoor position location system, and provide location service for user. In the first part, we will collect the RSS information of reference point to train the hybrid neural network models, and input the RSS information of track object to the model, and the model will provide the location of track object according to the RSS information. In the second part, we will calculate the performance of each neural network models and their weights were modified according to performance of each neural network. We will test the accuracy of location system again, and will use this system for patient care, smart home, and smart space.

Many kinds of anti-diabetic drugs that doctors could prescribe because of pharmaceutical progressing. Our study uses Protégé to build an anti-diabetic drugs ontology and a patient data ontology. The anti-diabetic drugs ontology was used to store medicine knowledge and the patient data ontology to store personal information. In this paper, we adopted the “Medical Guidelines for Clinical Practice for The Management of Diabetes Mellitus” provided by American Association of Clinical Endocrinologists with the help of an endocrinologist's help in Taichung hospital, in Taiwan. The nature, class, contraindications, and side effects of the drugs were collated and analyzed completely to build an ontology. We used Semantic Web Rule Language (SWRL) to build rules related to anti-diabetic drugs. The ontology knowledge was transformed to a suit format for reasoning by the Reasoner. Finally Java Expert Systems Shell (JESS) was used for reasoning, and the appropriate anti-diabetic drugs were recommended with the information about what need to be monitored, contraindications and side effects. In the experiments, the system shows that SWRL and JESS were combined to analyze the conditions of the diabetic, and then the most appropriate drugs were suggested.

Many FAQ (Frequently Asked Questions) systems have been proposed on the Internet. Users can easily obtain the information what they need, but they still have to read and organize those documents by themselves. However, they read the same contents from website in spite of men or women, elders or children. For different patients of diabetes, if a FAQ system can return personal health nursing documents that will be better. According the Department of Health Executive Yuan R.O.C. (Taiwan) to the recent statistics, most of patients lack chronic disease knowledge (e.g. diabetes, hypertension). Hospitals preach patients the knowledge of taking care of chronic disease. The knowledge can be shown by a narrator of the medical personnel, books, website to quickly help patients. However, the return of FAQ website information is fix to the users' questions. In this paper, we will set up a FAQ system based on domain ontology for diabetes health education documents and use FFCA (Fuzzy Formal Concept Analysis) method to help diabetic patients to get education documents. The primary experiments prove that our method is preciously to get users requirement documents.

The main function of IDS (intrusion detection system) is to protect the system, analyze and predict the behaviors of users. Then these behaviors will be considered an attack or a normal behavior. Though IDS has been developed for many years, the large number of return alert messages makes managers maintain system inefficiently. In this paper, we use RST (Rough Set Theory) and SVM (Support Vector Machine) to detect intrusions. First, RST is used to preprocess the data and reduce the dimensions. Next, the features selected by RST will be sent to SVM model to learn and test respectively. The method is effective to decrease the space density of data. The experiments will compare the results with different methods and show RST and SVM schema could improve the false positive rate and accuracy.

Car navigation systems are now widely used as a component for intelligent transportation systems. Route planning is the most important task of car navigation systems. Though modern car navigation systems incorporate various road information, even dynamic information, to generate optimal route, but they are yet to present routes according to driver's requirements or preferences. Most of the car navigation systems present a single best route or alternate routes according to systems predefined choices which may not satisfy the driver. In this work a fuzzy genetic approach is proposed to generate alternate routes according to driver's requirement and choice with fine tuning by using feed back mechanism. A simple simulation experiment proves the effectiveness and importance of the concept for developing more user friendly car navigation system with an outline of implementation.

One of the challenges to secure wireless sensor networks (WSNs) is to design secure pair-wise key agreement between any pair of resources-limited sensors while keeping strong connectivity. In this paper, we point out the fatal security weaknesses of Cheng-Agrawal's pair-wise key agreement scheme for WSNs, and then propose a new scheme. Compared to the existing schemes, the proposed scheme owns two outstanding merits- assurance of connectivity between any pair of nodes and security robustness against node capture attack.

In this paper, we will propose a domain ontology extensible method which can insert new keywords into the corresponding constructed domain ontology. The novel method uses TF-IDF (Term Frequency - Inverse Document Frequency) and LSA (Latent Semantic Analysis) to strengthen the semantic characteristic of keywords and transform the LSA matrix into a high dimensional space based on collected web pages. The similarity between keywords and concepts is evaluated by comparing the keywords on high dimensional matrix to the corresponding constructed ontology group center and individual concept. The keyword which has the highest similarity with the domain concept will be inserted into the offspring of the concept on the domain ontology and will become a new domain concept. The primary experiment uses the existing Chinese basketball ontology to examine its precision and the experimental results indicate that our proposed method has 73% accuracy.

Abstract: Research on semantic webs has become increasingly widespread in the computer science community. The core technology of a semantic web is an artefact called an ontology. The major problem in constructing an ontology is the long period of time required. Another problem is the large number of possible meanings for the knowledge in the ontology. In this paper, we present a novel ontology construction based on artificial neural networks and a Bayesian network. First, we collected web pages related to the problem domain using search engines. The system then used the labels of the HTML tags to select keywords, and used WordNet to determine the meaningful keywords, called terms. Next, it calculated the entropy value to determine the weight of the terms. After the above steps, the projective adaptive resonance theory neural network clustered the collected web pages and found the representative term of each cluster of web pages using the entropy value. The system then used a Bayesian network to insert the terms and complete the hierarchy of the ontology. Finally, the system used a resource description framework to store and express the ontology results.

New technologies are continuously being applied to high-speed networks such as distribution systems, on-line video-services, and some other applications. These services must ensure stable QoS (quality of service) and provide acceptable link costs, time delay, bandwidth and pack loss constraints, essentially a multicast routing problem. Computing such multicast routing problems is an NP-complete problem. In this paper, we propose a genetic algorithm using fuzzy selection (GFS) to address the problem of QoS under delay constraints. The GFS algorithm finds a multicast tree with minimum cost under delay constraints..

Almost all form documents contain line segments. We propose an efficient method to recognize the form document that contains at least one line segment. Our method is based on an efficient representation model of the form. The representation model uses three types of line segments to represent a form. All line segments are normalized and sorted after they are extracted. The normalization and sorting not only solve the problems of magnification and contraction but also provide a unified and efficient way of matching between forms. To make the recognition method more robust, fuzzy matching is used. Experimental results show the effectiveness and efficiency of the method.

A new approach is proposed to segment off-line handwritten Chinese characters. Many papers have been published on the off-line recognition of Chinese characters, and almost all of them focus on the recognition of isolated Chinese characters. The segmentation of text into characters was rarely discussed. The segmentation is an important preprocess of the off-line Chinese character recognition because correct recognition of characters relies on correct segmentation of characters. In handwritten Chinese characters, characters may be written to touch each other or to overlap with each other, therefore, the segmentation problem is not an easy one. In this paper, we present a novel method which uses strokes to build stroke bounding boxes first. Then, the knowledge-based merging operations are used to merge those stroke bounding boxes and finally, a dynamic programming method is applied to find the best segmentation boundaries. A series of experiments show that our method is very effective for off-line handwritten Chinese character segmentation.