This thesis investigates the application. of fuzzy logic techniques and user modelling to the process of information retrieval (IR) from the World Wide Web (WWW). The research issue is whether this process can be improved through such an application. The exponential rise of information itself as an invaluable global commodity, coupled with .acceierating development in. computing and telecommunications, and boosted by networked information sources such as the WWW, has led to the development of tools, such as search engines, to facilitate information search and retrieval. However, despite their sophistication, they are unable effectively to. address users' information. needs. Also, as the-WWW can be seen as a dynamic, continuously changing global information corpus, these tools suffer from the problems of irrelevancy and redundancy. Therefore, in order to overcome these problems and remain effective, IR systems need to become 'intelligent' in some way. It is from this premise that the focus of this research has developed. Initially, theoretical and investigative research into the areas ofIR from electronic sources and the nature of the Internet (including the WWW) revealed that highly sophisticated systems are being developed and there is a drive towards the integration of, for example, electronic libraries, COROM networks, and the WWW. Research into intelligent IR, the use of AI techniques to improve the IR process, informed an evaluation of various approaches. This revealed that a munber of techniques, for example, expert systems, neural networks and semantic networks, have been employed, with limited success. Owing to the nature of the WWW, though, many of the previous AI approaches are inapplicable as they rely too much on extensive knowledge of the retrieval corpus. However, the evaluation suggested that fuzzy logic, with its inherent ability to capture partial knowledge within fuzzy sets, is a valid approach. User modelling research indicated that adaptive user stereotypes are a fruitful way to represent different types of user and their information need. Here, these stereotypes are represented as fuzzy sets, ensuring flexibility and adaptivity. The goal of the reported research. then, was not to. develop an 'intelligent agent' but to apply fuzzy logic techniques and user modelling to the process of user query formulation, in order to test the research issue. This issue was whether the application of these techniques could improve the IR process. A prototype system, the Fuzzy Modelling Query Assistant (FMQA), was developed that attempts intelligently to assist the user in capturing their information need. The concept was to refine the user's query before submitting it to an existing search engine, in order to improve upon the IR results of using the search tool alone. To address the research issue, a user study of the FMQA was performed. The design and conduct is reported in depth. The study results were analysed and the findings are given. The results indicate that,. for certain types of user especially, the FMQA does provide improvement in the IR process, in terms of the results. There is a critical review of the research aims in the light of the results, conclusions are drawn and recommendations for future research given.

This thesis is an effort to develop closed-loop control strategies on computer networks and study their stability in the presence of network delays and packet losses. An algorithm using predictors was designed to ensure the system stability in presence of network delays and packet losses. A single actuator magnetic ball levitation system was used as a test bed to validate the proposed algorithm. A brief study of real-time requirements of the networked control system is presented and a client-server architecture is developed using real-time operating environment to implement the proposed algorithm. Real-time performance of the communication on Ethernet based on user datagram protocol (UDP) was explored and UDP is presented as a suitable protocol for networked control systems. Predictors were designed based on parametric estimation models. Autoregressive (AR) and autoregressive moving average (ARMA) models of various orders were designed using MATLAB and an eighth order AR model was adopted based on the best-fit criterion. The system output was predicted several steps ahead using these predictors and control output was calculated using the predictions. This control output output was used in the events of excessive network delays to maintain system stability. Experiments employing simulations of consecutive packet losses and network delays were performed to validate the satisfactory performance of the predictor based algorithm. The current system compensates for up to 20 percent data losses in the network without loosing stability.

A networked control system (NCS) is a feedback control system that has its control loop physically connected via real-time communication networks. To meet the demands of `teleautomation', modularity, integrated diagnostics, quick maintenance and decentralization of control, NCSs have received remarkable attention worldwide during the past decade. Yet despite their distinct advantages, NCSs are suffering from network-induced constraints such as time delays and packet dropouts, which may degrade system performance. Therefore, the network-induced constraints should be incorporated into the control design and related studies.<p> For the problem of state estimation in a network environment, we present the strategy of simultaneous input and state estimation to compensate for the effects of unknown input missing. A sub-optimal algorithm is proposed, and the stability properties are proven by analyzing the solution of a Riccati-like equation.<p> Despite its importance, system identification in a network environment has been studied poorly before. To identify the parameters of a system in a network environment, we modify the classical Kalman filter to obtain an algorithm that is capable of handling missing output data caused by the network medium. Convergence properties of the algorithm are established under the stochastic framework.<p> We further develop an adaptive control scheme for networked systems. By employing the proposed output estimator and parameter estimator, the designed adaptive control can track the expected signal. Rigorous convergence analysis of the scheme is performed under the stochastic framework as well.

A networked control system (NCS) is a feedback control system that has its control loop physically connected via real-time communication networks. To meet the demands of `teleautomation', modularity, integrated diagnostics, quick maintenance and decentralization of control, NCSs have received remarkable attention worldwide during the past decade. Yet despite their distinct advantages, NCSs are suffering from network-induced constraints such as time delays and packet dropouts, which may degrade system performance. Therefore, the network-induced constraints should be incorporated into the control design and related studies.<p> For the problem of state estimation in a network environment, we present the strategy of simultaneous input and state estimation to compensate for the effects of unknown input missing. A sub-optimal algorithm is proposed, and the stability properties are proven by analyzing the solution of a Riccati-like equation.<p> Despite its importance, system identification in a network environment has been studied poorly before. To identify the parameters of a system in a network environment, we modify the classical Kalman filter to obtain an algorithm that is capable of handling missing output data caused by the network medium. Convergence properties of the algorithm are established under the stochastic framework.<p> We further develop an adaptive control scheme for networked systems. By employing the proposed output estimator and parameter estimator, the designed adaptive control can track the expected signal. Rigorous convergence analysis of the scheme is performed under the stochastic framework as well.

ABSTRACT The small & median enterprises that possess elasticity and efficiency always play to very important roles in the economic development of Taiwan. However the small & median enterprises hold technology and R&D ability in start-up stage, which are more fragile and to wait their maturity. Beside their business scales, markets, capital and human relationship networks are less. They face to be high uncertainty and market competition from start-up to firm business all the time. They are very careful under these situations, but still fail after all. Incubators are sponsors help entrepreneurs or start-ups that have originality, technique and potentiality to give some supports in early stage that lacks experiences, capital, resources and reputation. In order to achieve diversification purposes such as job creation, assisting the economic revitalization of an area, creating wealth, fostering new product/enterprise and so on. We expect from development experience of foreign incubators to base on " an advice from others may help one's defects" viewpoint. To conclude a successful business model of incubator through researches of relevant literature and depth interviews of experts. Our discoveries of researches as follow, ¡¯The importance of incubator successful business model take turns¡G 1.Clearing business missions/purposes. 2.Recruiting excellent managers who can run an incubator to regard as a business. 3.Offering services and helps for tenants. 4.Drawing up a feasibility/business plan that can attract investors. 5.The board of directors that can promise missions/purposes and help to elevate competence. 6.Maintaining 3 to 5 years finance. 7.Obtaining helps and supports of stakeholders, further develop to become a networked incubator. Furthermore the supports and participation of school decision makers are very important because the generalities are academic incubators in domestic. ¡¯The best important item of incubator successful business model in individual condition discriminates¡G 1.Clearing business missions/purposes.¡G"commercialize key technology" 2. Drawing up a feasibility/business plan that can attract investors.¡G"strategy" 3. The board of directors that can promise missions/purposes and help to elevate competence.¡G"accumulated resources in past practice experiences" 4.Recruiting excellent managers who can run an incubator to regard as a business.¡G"entrepreneurship" 5.Offering services and helps for tenants.¡G" technology & talent support" 6.Tenants' entrance criteria¡G"demonstrated business capability of principals" 7.Maintaining 3 to 5 years finance.¡G"self-sufficiency in a proper way " 8.Obtaining helps and supports of stakeholders.¡G"industries" 9.Developing key of a networked incubator.¡G" preferential access to potential resources and partners than competitor from network." Suggestions¡G 1.Creating incentives in order to conduct continuously for academic incubators and professors and managers. 2.Local governments should play more aggressive roles for the development of incubators, and integrate a distinguishing feature of universities and local industries. 3.To encourage development and collaboration of private incubators through governments' reward and taxes reduction. 4.The incubation objects should reinforce to foster traditional businesses that have upgrade and transfer pattern wishes except start-ups. 5.Incubators should promote incubation outcomes to internal and external market through internet or trade organizations. 6.The institutions or organizations of abroad successful incubator are invited by Small & Median Enterprises Development Fund to guide or transfer technologies.

This thesis is an e&#64256;ort to develop closed-loop control strategies on computer networks and study their stability in the presence of network delays and packet losses. An algorithm using predictors was designed to ensure the system stability in presence of network delays and packet losses. A single actuator magnetic ball levitation system was used as a test bed to validate the proposed algorithm. A brief study of real-time requirements of the networked control system is presented and a client-server architecture is developed using real-time operating environment to implement the proposed algorithm. Real-time performance of the communication on Ethernet based on user datagram protocol (UDP) was explored and UDP is presented as a suitable protocol for networked control systems. Predictors were designed based on parametric estimation models. Autoregressive (AR) and autoregressive moving average (ARMA) models of various orders were designed using MATLAB and an eighth order AR model was adopted based on the best-&#64257;t criterion. The system output was predicted several steps ahead using these predictors and control output was calculated using the predictions. This control output output was used in the events of excessive network delays to maintain system stability. Experiments employing simulations of consecutive packet losses and network delays were performed to validate the satisfactory performance of the predictor based algorithm. The current system compensates for up to 20 percent data losses in the network without loosing stability.

Η παρούσα διδακτορική διατριβή στοχεύει στην μελέτη και ανάλυση μεταβλητών συστημάτων με απώτερο σκοπό την ανάπτυξη αλγορίθμων στο χώρο της αναγνώρισης συστημάτων, πρόβλεψης εξόδου αυτών, και δημιουργίας προσαρμοστικών ελεγκτών που θα επιδεικνύουν σθεναρότητα απέναντι στις μεταβολές των υπό μελέτη συστημάτων. Η μεταβολή στα συστήματα αυτά προκύπτει κυρίως από την εισαγωγή τυχαίων χρονικών καθυστερήσεων στην είσοδο και στην έξοδο αυτών. Οι καθυστερήσεις αυτές, είναι δυνατόν να είναι σταθερές ή μεταβαλλόμενες, γνωστές ή άγνωστες, ντετερμινιστικές ή στοχαστικές. Σαν τελικό αποτέλεσμα οι εισαγόμενες χρονοκαθυστερήσεις είναι δυνατόν να αλλάξουν την συμπεριφορά του συστήματος οδηγώντας αυτό στην αστάθεια. Στην παρούσα διδακτορική διατριβή προτείνονται μέθοδοι μοντελοποίησης αυτών των χρονοκαθυστερήσεων σε συνδυασμό με τους αλγορίθμους αναγνώρισης και πρόβλεψης εξόδου συστημάτων υποκείμενων σε αυτές τις χρονικές καθυστερήσεις. Επιπροσθέτως παρατίθενται πρότυπες και κατάλληλα διαμορφωμένες δομές προσαρμοστικού ελέγχου και μελέτης της ευστάθειας αυτών των συστημάτων. / The aim of this dissertation is the study and analysis of varying systems, as also the development of algorithms in the field of: a) system identification, b) system prediction, and c) adaptive control that will be able to be robust to the variations of the systems under study. These variations in these systems are induced mainly from the insertion of time delays in the input and output of those systems. These delays can be constant, or varying, known or unknown, deterministic or stochastic. As a final result these time delays can alter the behavior and the performance of the system while driving it to instability. In this dissertation new algorithms are proposed for the modeling of these time delays in conjunction with the algorithms for the identification and output prediction of systems with time delays. Additionally new algorithms of adaptive control and stability analysis of the controlled systems are proposed.

Wireless networked control systems use shared wireless links to communicate between sensors and controllers, and require a channel access policy to arbitrate access to the links. Existing multiple access protocols perform this role in an agnostic manner, by remaining insular to the applications that run over the network. This approach does not give satisfactory control performance guarantees. To enable the use of wireless networks in emerging industrial applications, we must be able to systematically design wireless networked control systems that provide guaranteed performances in resource-constrained networks. In this thesis, we advocate the use of state-based channel access policies. A state-based policy uses the state of the controlled plant to influence access to the network. The state contains information about not only the plant, but also the network, due to the feedback in the system. Thus, by using the state to decide when and how frequently to transmit, a control system can adapt its contribution to the network traffic, and enable the network to adapt access to the plant state. We show that such an approach can provide better performance than existing methods. We examine two different state-based approaches that are distributed and easy to implement on wireless devices: event-based scheduling and adaptive prioritization. Our first approach uses events to reduce the traffic in the network. We use a state-based scheduler in every plant sensor to generate non-coordinated channel access requests by selecting a few critical data packets, or events, for transmission. The network uses a contention resolution mechanism to deal with simultaneous channel access requests. We present three main contributions for this formulation. The first contribution is a structural analysis of stochastic event-based systems, where we identify a dual predictor architecture that results in separation in design of the state-based scheduler, observer and controller. The second contribution is a Markov model that describes the interactions in a network of event-based systems. The third contribution is an analysis of the stability of event-based systems, leading to a stabilizing design of event-based policies. Our second approach uses state-based priorities to determine access to the network. We use a dominance protocol to evaluate priorities in a contention-based setting, and characterize the resulting control performance. An implementation and evaluation of this channel access mechanism on sensor nodes is also presented. The thesis finally examines the general networked control problem of jointly optimizing measurement and control policies, when a nonlinear measurement policy is used to perform quantization, event-triggering or companding. This contribution focuses on some of the fundamental aspects of analyzing and synthesizing control systems with state-based measurement policies in a more generalized setting. We comment on the dual effect, certainty equivalence and separation properties for this problem. In particular, we show that it is optimal to apply separation and certainty equivalence to a design problem that permits a dynamic choice of the measurement and control policies. / <p>QC 20140408</p>

This study evaluated the third design iteration of a networked teacher professional development (nTPD) implementation. In particular, the study explored the kinds of teacher technology professional learning that resulted as a consequence of nTPD participation. As part of an ongoing design-based research program, the goal of this study was to evaluate the teacher learning resulting from participation in online-delivered TPD activities. In addition the results inform an evolving model of nTPD articulating the components and elements of the online learning activities that have value in supporting and/or advancing teacher practices. The results of this study indicate that teachers who participate in nTPD find the experiential learning activities and the sharing of resources and lesson plans to be valuable for their professional practice. NTPD, delivered in a social networking site environment, results in new kinds of teacher learning opportunities. Some of these new learning opportunities include shared digital curation activities and unique cognitive-apprenticeship type activities described further as “learning over the shoulders of giants.” In theory, nTPD provides teachers with opportunities to connect with others who are teaching in similar curricular areas to identify, develop, and share resources that can support their teaching practice. In practice, the articulation of a revised nTPD model and design principles provides developers of online-networked TPD with guidelines for the development of valued learning activities, particularly for technology TPD topics. / 2013-03

Management of Collaborative Networked Organizations faces various challenges in terms of decision-making. Particularly, in complex and multi-player environments, like healthcare, it is not easy to find the roots of low performance processes, and unmet goals. This research provides a framework, as well as associated techniques to analyze the value network, identify problematic actors, and consequently, find the best possible solution to change them. The proposed framework consists of two main components: Analyzing the value network, and Multi-Criteria Decision Making. To analyze the value network of a collaborative environment, in addition to the existing techniques, four complementary components are introduced: Actors’ value interchanges matrix, Value Gantt chart, Identifying problematic actors flowchart, and Actors’ ease of substitution table. Employing these hybrid analyses, decision makers gain a better understanding of the bottlenecks in the value network, current conditions and contributions of the involved actors, and the consequences of considering various alternatives. Then, by applying one of the Multi-Criteria Decision Making methods, and based on pre-defined criteria, possible alternatives are analyzed and outlined. As a proof of concept and validation of the proposed methods, we reviewed a scenario of patient flow and wait times in healthcare. We derive the value network for collaborative processes in a hospital, specify the roles’ of actors, identify the bottlenecks, then rank the solutions, and suggest possible changes to improve the performance of the collaborative environment.