Browsing by Author "Lin, Szu-Yin"

Now showing 1 - 15 of 15

A Convolutional Neural Network Approach for Dental Panoramic Radiographs Classification
(2017-12-01)
Kuo, Yu-Fang
;
Lin, Szu-Yin
;
Wu, Calvin H.
;
Chen, Shih-Lun
;
Lin, Ting-Lan
Dental Panoramic Radiograph (DPR) has been widely accepted in the dental industry in recent years for providing valuable information to dental practitioners. Conventional fashion of analyzing this information has always been relying on the professional individuals. Nonetheless, the recent technology breakthrough with computer vision is promising in reducing effort and labor costs. The purpose of this study is to propose an automated DPR classification method that could be used to identify condition such as having undergone endodontic therapy. This new method used a combination of unique preprocessing procedure with highly developed image recognition algorithm known as Convolutional Neural Network (CNN). The new method firstly preprocesses and normalizes the DPR images as the training dataset, then trains the network model using the training data. Finally, testing datasets will then be used for generating evaluation results. Based from the experimental results, the trained network with preprocessed training datasets performed an accuracy rate of 85.3%. In addition, the performance of the network could still be improved by increasing the number of training datasets and using more preprocessing methods. In conclusion, computer visions can provide confident better performance in DPR classification in practices such as DPR classification.
A Cost-Effective Planning Graph Approach for Large-Scale Web Service Composition
(2012)
Lin, Szu-Yin
;
Lin, Guan-Ting
;
Chao, Kuo-Ming
;
Lo, Chi-Chun
;
資訊管理與財務金融系註：原資管所+財金所
Web Service Composition (WSC) problems can be considered as a service matching problem, which means that the output parameters of a Web service can be used as inputs of another one. However, when a very large number of Web services are deployed in the environment, the service composition has become sophisticated and complicated process. In this study, we proposed a novel cost-effective Web service composition mechanism. It utilizes planning graph based on backward search algorithm to find multiple feasible solutions and recommends a best composition solution according to the lowest service cost. In other words, the proposed approach is a goal-driven mechanism, which can recommend the approximate solutions, but it consumes fewer amounts of Web services and less nested levels of composite service. Finally, we implement a simulation platform to validate the proposed cost-effective planning graph mechanism in large-scale Web services environment. The simulation results show that our proposed algorithm based on the backward planning graph has reduced by 94% service cost in three different environments of service composition that is compared with other existing service composition approaches which are based on a forward planning graph.
A dynamic data driven-based semi-distributed reputation mechanism in unknown networks
(2015-10-01)
Lin, Szu-Yin
;
Chou, Ping-Hsien
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Trust is a crucial concern related to unknown networks. A mechanism that distinguishes trustworthy and untrustworthy nodes is essential. The effectiveness of the mechanism depends on the accuracy of a node\'s reputation. The dynamics of trust often occurs in a trusted network and causes intoxication and disguises of the nodes, resulting in abnormal behaviors. This study proposes a semi-distributed reputation mechanism based on a dynamic data-driven application system. This mechanism includes two reputations, local reputation (LRep) and global reputation (GRep). LRep is dynamically and selectively injected into a central controller, and this controller collects the injected data to compute GRep, which contains the neural network prediction method, and returns it to provide reference to the distributed nodes. The proposed mechanism focuses on dynamics of trust and the balance between distributed nodes and the central controller. Experimental results showed that GRep was computable with only 52.21% (average) LReps upload and that GRep increased or reduced by 26.5% (average) in a short period, demonstrating that the proposed mechanism effectively handles the problem of dynamics of trust. (C) 2015 Elsevier B.V. All rights reserved.
A Dynamic Data-Driven Fine-Tuning Approach for Stacked Auto-Encoder Neural Network
(2017-01-01)
Lin, Szu-Yin
;
Chiang, Chi-Chun
;
Hung, Zih-Siang
;
Zou, Yu-Hui
;
資訊管理與財務金融系註：原資管所+財金所
With the advent of the big data era, dynamic and real-time data have increased in both volume and varieties. It is a difficult task to achieve an accurate prediction results to rapidly dynamic changing data. The stacked auto-encoder is a neural network approach in machine learning for feature extraction. It attempts to model high-level abstractions and dimension reduction in data by using multiple processing layers. However, some of the common issues may occur during the implementation process of deep learning or neural network, such as input data having over-complicated dimension, and unable to execute in a dynamic environment. Therefore, it will be helpful if we combine dynamic data-driven concept with stacked auto-encoder neural network to obtain the dynamic data correlation or relationship between prediction results and actual data in a dynamic environment. This study applies the concept of dynamic data-driven to obtain the correlations between the prediction goals and numbers of different combination results. The methods of association analysis, sequence analysis, and stacked auto-encoder neural network are applied to design a dynamic data-driven system based on deep learning.
A Semi-Distributed Reputation Mechanism based on Dynamic Data-Driven Application System
(2014-01-01)
Lin, Szu-Yin
;
Chou, Ping-Hsien
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Trust is one of the important issues related to unknown networks. A mechanism which can distinguish a trustworthy node from an untrustworthy one is essential. The effectiveness of the mechanism depends on the accuracy of node\'s reputation. Dynamics of Trust often happens in a trusted network. It causes intoxication and disguise for nodes, resulting in abnormal behaviors. This paper proposes a semi-distributed reputation mechanism based on Dynamic Data-Driven Application System. It focuses on the Dynamics of Trust and the balance between Distributed Nodes and the Central Controller. The experimental results show that the proposed mechanism upload only averages 52.21% of the data to compare with all of uploads. It can also effectively handle the problem of Dynamics of Trust.
A Social Trust based Recommendation Mechanism for Web Service Dynamic Collaboration
(2013)
Lin, Szu-Yin
;
Yang, Ya-Ching
;
Lo, Chi-Chun
;
Chao, Kuo-Ming
;
資訊管理與財務金融系註：原資管所+財金所
In the distributed commercial environment, many enterprises utilize Web services to provide their resources such as data and applications to consumers. They collaborate with other relevant enterprises to accomplish tasks by sharing processes and resources. However, it is difficult to find the appropriate web service combination for enterprise, because the Quality of Services (QoS) that are widely used to select Web services is not sufficient anymore. In dynamic collaboration environment, it does not only need to consider the QoS, functionality and degree of matching, but also the social trust relationship between enterprises. This study proposes a social-trust-based recommendation mechanism for service-oriented dynamic collaboration. It uses the legal binding of Service Level Agreement and the characteristics of Social Trust Network of Web services to recommend proper and trustworthy services to support dynamic collaboration.
A trustworthy QoS-based collaborative filtering approach for web service discovery
(2014-07-01)
Lin, Szu-Yin
;
Lai, Chin-Hui
;
Wu, Chih-Heng
;
Lo, Chi-Chun
;
資訊管理與財務金融系註：原資管所+財金所
Many network services which process a large quantity of data and knowledge are available in the distributed network environment, and provide applications to users based on Service-Oriented Architecture (SOA) and Web services technology. Therefore, a useful web service discovery approach for data and knowledge discovery process in the complex network environment is a very significant issue. Using the traditional keyword-based search method, users find it difficult to choose the best web services from those with similar functionalities. In addition, in an untrustworthy real world environment, the QoS-based service discovery approach cannot verify the correctness of the web services\' Quality of Service (QoS) values, since such values guaranteed by a service provider are different from the real ones. This work proposes a trustworthy two-phase web service discovery mechanism based on QoS and collaborative filtering, which discovers and recommends the needed web services effectively for users in the distributed environment, and also solves the problem of services with incorrect QoS information. In the experiment, the theoretical analysis and simulation experiment results show that the proposed method can accurately recommend the needed services to users, and improve the recommendation quality. (C) 2014 Elsevier Inc. All rights reserved.
A Trustworthy QoS-based Mechanism for Web Service Discovery Based on Collaborative Filtering
(2013)
Lin, Szu-Yin
;
Lai, Chin-Hui
;
Wu, Chih-Heng
;
Lo, Chi-Chun
;
資訊管理與財務金融系註：原資管所+財金所
Many network services which process a large quantity of data and knowledge are available in the distributed network environment, and provide applications to users based on Service-Oriented Architecture (SOA) and Web services technology. Therefore, a useful web service discovery approach for data and knowledge discovery process in the complex network environment is a very significant issue. U sing the traditional keyword-based search method, users find it difficult to choose the best web services from those with similar functionalities. In addition, in an untrustworthy real world environment, the QoS-based service discovery approach cannot verify the correctness of the web services\' Quality of Service (QoS) values, since such values guaranteed by a service provider are different from the real ones. This work proposes a trustworthy two-phase web service discovery mechanism based on QoS and collaborative filtering, which discovers and recommends the needed web services effectively for users in the distributed environment, and also solves the problem of services with incorrect QoS information. In the experiment, the theoretical analysis and simulation experiment results show that the proposed method can accurately recommend the needed services to users, and improve the recommendation quality.
Automated knowledge discovery and semantic annotation for network and web services
(2016-07-01)
Lin, Szu-Yin
;
Chung, Chia-Chen
;
Hu, Wei-Che
;
Hung, Chihli
;
Chen, Shih-Lun
With the rise of the Internet of things, the smart environmental issue is becoming increasingly important. Sensor web is one of the best solutions to this issue and provides the advantages of sensor networks and web services. Ontology web language for services (OWL-S) is an OWL-based web services ontology, which provides the ability to describe the semantics of web services and their capabilities in a formal and machine-processable manner. Moreover, it aids semantic service matching, selection and composition. However, automatically annotating semantic web services is a highly complicated and tedious task. In this study, we propose a methodology to uncover information in the history data and profiles of web services and then semantically annotate them. With the proposed approach, semantic relationships between web services could be extracted via a combination of association rules and input/ output matching. Our results show that this hybrid automated knowledge-discovery approach works better than traditional approaches do. We also provide a scenario to explain how the proposed methodology works.
Automatic Clustering and Semantic Annotation for Dynamic IoT Sensor Data
(2018-01-01)
Yu, Ching-Tzu
;
Zou, Yu-Hui
;
Li, Hao-Yu
;
Lin, Szu-Yin
;
資訊管理與財務金融系註：原資管所+財金所
In a dynamic IoT environment, distributed sensors are used to collect real-time data continually. However, it is difficult to transform the dynamic data into a machine-readable and machine-interpretable form. we propose a semantic annotation approach to annotate sensor data via semantics. Firstly, this approach builds an ontology based on Semantic Sensor Network Ontology (SSN Ontology) for dynamic IoT sensor data. Then, the new knowledge is collected from input data by using the K-Means clustering, and to update the semantic information into the base ontology. The updated ontology forms the basis for semantic annotation.
Automatic Knowledge Discovery and Semantic Annotation for Web Services
(2015)
Lin, Szu-Yin
;
Chung, Chia-Chen
;
Hu, Wei-Che
;
Hung, Chihli
;
資訊管理與財務金融系註：原資管所+財金所
OWL-S (formally DAML-S) is an OWL-based Web services ontology, which gives the ability for describing the semantics of Web services and their capabilities in a formal and machine-processable manner. Moreover, it is helpful to semantic services matching, selection, and composition. However, it is a very complicated and heavy task to annotate semantic Web services manually by human. In this paper, we proposed a methodology to discover knowledge from the history data and profiles of Web services, and then semantic annotate them automatically. With the proposed approach, semantic relations between Web services could be extracted by combining association rules and input/output matching. We also give a scenario to explain how the proposed methodology works.
Dynamic Data Driven-based Automatic Clustering and Semantic Annotation for Internet of Things Sensor Data
(2019-01-01)
Lin, Szu-Yin
;
Li, Jun-Bin
;
Yu, Ching-Tzu
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Faced with the advent of the era of smart Internet of Things (IoT), a large amount of sensor data and a large number of intelligent applications have been introduced into our lives. However, the dynamic and multimodal nature of data makes it challenging to transform them into machine-readable and machine-interpretable forms. In this study, a semantic annotation method is proposed to annotate sensor data through semantics. First, the method constructs an initial ontology based on the semantic sensor network (SSN) ontology for dynamic loT sensor data. Second, through K-means clustering, new knowledge is extracted from input data, and the semantic information is used for updating the initial ontology. The updated ontology then forms the basis of semantic annotation. In this study, an experiment is performed to analyze the data collected from sensors every 10 s for a period of one month. From the results of simulation experiments, we found useful knowledge from new data. With more available knowledge, sensor data can be annotated with higher adequacy.
Dynamic fine-tuning stacked auto-encoder neural network for weather forecast
(2018-12-01)
Lin, Szu-Yin
;
Chiang, Chi-Chun
;
Li, Jung-Bin
;
Hung, Zih-Siang
;
Chao, Kuo-Ming
With the advent of the big data era, dynamic and real-time data have increased in both volume and variety. It is difficult to make accurate predictions regarding data as they undergo rapid and dynamic changes. Autonomous cloud computing aims to reduce the time required for traditional machine learning. The stacked auto-encoder is a neural network approach in machine learning for feature extraction. It attempts to model high-level abstractions and to reduce data dimensions by using multiple processing layers. However, some common issues may occur during the implementation of deep learning or neural network models, such as over-complicated dimensions of the input data and difficulty in processing dynamic data. Therefore, combining the concept of dynamic data-driven system with a stacked auto-encoder neural network will help obtain the dynamic data correlation or relationship between the prediction results and actual data in a dynamic environment. This study applies the concept of a dynamic data-driven system to obtain the correlations between the prediction goals and number of different combination results. Association analysis, sequence analysis, and stacked auto-encoder neural network are employed to design a dynamic data-driven system based on deep learning. (C) 2018 Elsevier B.V. All rights reserved.
On the Design and Development of a Novel Real-Time Transaction Price Estimation System
(2012)
Chen, Chi-Hua
;
Lin, Szu-Yin
;
Chang, Hsu-Chia
;
Lo, Chi-Chun
;
資訊管理與財務金融系註：原資管所+財金所
In global open economy, transaction price estimation is an important issue in finance under stock price fluctuation. However, the transaction price is depend on the investors' expect price which is difficult to be captured in used and implemented models for transaction price estimation. Therefore, this paper proposes the Transaction Price Estimation Model (TPEM) and designs a novel Real-time Transaction Price Estimation System (RTPES) which adopts the TPEM based on the efficient market hypothesis to estimate the investors' expect price. In experiments, the simulation results show that the average accuracy of TPEM is 99.37%. This approach is feasible for stock investment decision-making.
Service-Oriented Dynamic Data Driven Application Systems to Traffic Signal Control
(2010)
Lin, Szu-Yin
;
Chao, Kuo-Ming
;
Lo, Chi-Chun
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
Dynamic traffic signal control in Intelligent Transportation Systems (ITS) has been widely developed in major cities' urban areas around the world to provide more efficient way for solving traffic congestion problem. However, the problems in dynamic traffic signal control systems such as system flexibility, data standard common interface, transmission of required information, prediction performance, and real-time measurement data are all important issues. Therefore, in this paper, we proposed a new framework to integrate dynamic data driven application systems (DDDAS) with service-oriented architecture (SOA) and web services technology, for a decentralized traffic signal control system. An efficient and effective distributed data-driven optimization algorithm is designed in this framework to produce an optimal result and best strategy for signal control.