53 resultados para expert systems
Resumo:
The influence of IT investment on hospital efficiency and quality are of great interest to healthcare executives as well as insurers. Few studies have examined how IT investments influence both efficiency and quality or whether there is an optimal IT investment level that influences both in the desired direction. Decision makers in healthcare wonder if there are tradeoffs between their pursuit of hospital operational efficiency and quality. Our study involving a 2-stage double bootstrap DEA analysis of 187 US hospitals over 2. years found direct effects of IT investment upon service quality and a moderating effect of quality upon operational efficiency. Further, our findings indicate a U-shaped relationship between IT investments and operational efficiency suggesting that IT investments have diminishing returns beyond a certain point.
Resumo:
Recommender system is a specific type of intelligent systems, which exploits historical user ratings on items and/or auxiliary information to make recommendations on items to the users. It plays a critical role in a wide range of online shopping, e-commercial services and social networking applications. Collaborative filtering (CF) is the most popular approaches used for recommender systems, but it suffers from complete cold start (CCS) problem where no rating record are available and incomplete cold start (ICS) problem where only a small number of rating records are available for some new items or users in the system. In this paper, we propose two recommendation models to solve the CCS and ICS problems for new items, which are based on a framework of tightly coupled CF approach and deep learning neural network. A specific deep neural network SADE is used to extract the content features of the items. The state of the art CF model, timeSVD++, which models and utilizes temporal dynamics of user preferences and item features, is modified to take the content features into prediction of ratings for cold start items. Extensive experiments on a large Netflix rating dataset of movies are performed, which show that our proposed recommendation models largely outperform the baseline models for rating prediction of cold start items. The two proposed recommendation models are also evaluated and compared on ICS items, and a flexible scheme of model retraining and switching is proposed to deal with the transition of items from cold start to non-cold start status. The experiment results on Netflix movie recommendation show the tight coupling of CF approach and deep learning neural network is feasible and very effective for cold start item recommendation. The design is general and can be applied to many other recommender systems for online shopping and social networking applications. The solution of cold start item problem can largely improve user experience and trust of recommender systems, and effectively promote cold start items.
Resumo:
Different types of sentences express sentiment in very different ways. Traditional sentence-level sentiment classification research focuses on one-technique-fits-all solution or only centers on one special type of sentences. In this paper, we propose a divide-and-conquer approach which first classifies sentences into different types, then performs sentiment analysis separately on sentences from each type. Specifically, we find that sentences tend to be more complex if they contain more sentiment targets. Thus, we propose to first apply a neural network based sequence model to classify opinionated sentences into three types according to the number of targets appeared in a sentence. Each group of sentences is then fed into a one-dimensional convolutional neural network separately for sentiment classification. Our approach has been evaluated on four sentiment classification datasets and compared with a wide range of baselines. Experimental results show that: (1) sentence type classification can improve the performance of sentence-level sentiment analysis; (2) the proposed approach achieves state-of-the-art results on several benchmarking datasets.
Resumo:
INTRODUCTION: Liposomes remain at the forefront of drug and vaccine design owing to their well-documented abilities to act as delivery vehicles. Nevertheless, the concept of liposomes as delivery vehicles is not a new one, with most works focusing on their use for the delivery of genes and drugs. However, in the last 10 years a significant amount of research has focused on using liposomes as vaccine adjuvants, not only as an antigen delivery vehicle but also as a tool to increase the immunogenicity of peptide and protein antigens. AREAS COVERED: This paper reviews liposomal adjuvants now in vaccine development, with particular emphasis on their adjuvant mechanism and how specific physicochemical characteristics of liposomes affect the immune response. The inclusion of immunomodulators is also discussed, with prominence given to Toll-like receptor ligands. EXPERT OPINION: The use of liposomes as vaccine delivery systems is evolving rapidly owing to the combined increase in technological advances and understanding of the immune system. Liposomes that contain and deliver immunostimulators and antigens are now being developed to target diseases that require stimulation of both humoral and cell-mediated immune responses. The CAF liposomal system, described in detail in this review, is one liposomal model that shows such flexibility.
Resumo:
Classification of metamorphic rocks is normally carried out using a poorly defined, subjective classification scheme making this an area in which many undergraduate geologists experience difficulties. An expert system to assist in such classification is presented which is capable of classifying rocks and also giving further details about a particular rock type. A mixed knowledge representation is used with frame, semantic and production rule systems available. Classification in the domain requires that different facets of a rock be classified. To implement this, rocks are represented by 'context' frames with slots representing each facet. Slots are satisfied by calling a pre-defined ruleset to carry out the necessary inference. The inference is handled by an interpreter which uses a dependency graph representation for the propagation of evidence. Uncertainty is handled by the system using a combination of the MYCIN certainty factor system and the Dempster-Shafer range mechanism. This allows for positive and negative reasoning, with rules capable of representing necessity and sufficiency of evidence, whilst also allowing the implementation of an alpha-beta pruning algorithm to guide question selection during inference. The system also utilizes a semantic net type structure to allow the expert to encode simple relationships between terms enabling rules to be written with a sensible level of abstraction. Using frames to represent rock types where subclassification is possible allows the knowledge base to be built in a modular fashion with subclassification frames only defined once the higher level of classification is functioning. Rulesets can similarly be added in modular fashion with the individual rules being essentially declarative allowing for simple updating and maintenance. The knowledge base so far developed for metamorphic classification serves to demonstrate the performance of the interpreter design whilst also moving some way towards providing a useful assistant to the non-expert metamorphic petrologist. The system demonstrates the possibilities for a fully developed knowledge base to handle the classification of igneous, sedimentary and metamorphic rocks. The current knowledge base and interpreter have been evaluated by potential users and experts. The results of the evaluation show that the system performs to an acceptable level and should be of use as a tool for both undergraduates and researchers from outside the metamorphic petrography field. .
Resumo:
The work described was carried out as part of a collaborative Alvey software engineering project (project number SE057). The project collaborators were the Inter-Disciplinary Higher Degrees Scheme of the University of Aston in Birmingham, BIS Applied Systems Ltd. (BIS) and the British Steel Corporation. The aim of the project was to investigate the potential application of knowledge-based systems (KBSs) to the design of commercial data processing (DP) systems. The work was primarily concerned with BIS's Structured Systems Design (SSD) methodology for DP systems development and how users of this methodology could be supported using KBS tools. The problems encountered by users of SSD are discussed and potential forms of computer-based support for inexpert designers are identified. The architecture for a support environment for SSD is proposed based on the integration of KBS and non-KBS tools for individual design tasks within SSD - The Intellipse system. The Intellipse system has two modes of operation - Advisor and Designer. The design, implementation and user-evaluation of Advisor are discussed. The results of a Designer feasibility study, the aim of which was to analyse major design tasks in SSD to assess their suitability for KBS support, are reported. The potential role of KBS tools in the domain of database design is discussed. The project involved extensive knowledge engineering sessions with expert DP systems designers. Some practical lessons in relation to KBS development are derived from this experience. The nature of the expertise possessed by expert designers is discussed. The need for operational KBSs to be built to the same standards as other commercial and industrial software is identified. A comparison between current KBS and conventional DP systems development is made. On the basis of this analysis, a structured development method for KBSs in proposed - the POLITE model. Some initial results of applying this method to KBS development are discussed. Several areas for further research and development are identified.
Resumo:
The present scarcity of operational knowledge-based systems (KBS) has been attributed, in part, to an inadequate consideration shown to user interface design during development. From a human factors perspective the problem has stemmed from an overall lack of user-centred design principles. Consequently the integration of human factors principles and techniques is seen as a necessary and important precursor to ensuring the implementation of KBS which are useful to, and usable by, the end-users for whom they are intended. Focussing upon KBS work taking place within commercial and industrial environments, this research set out to assess both the extent to which human factors support was presently being utilised within development, and the future path for human factors integration. The assessment consisted of interviews conducted with a number of commercial and industrial organisations involved in KBS development; and a set of three detailed case studies of individual KBS projects. Two of the studies were carried out within a collaborative Alvey project, involving the Interdisciplinary Higher Degrees Scheme (IHD) at the University of Aston in Birmingham, BIS Applied Systems Ltd (BIS), and the British Steel Corporation. This project, which had provided the initial basis and funding for the research, was concerned with the application of KBS to the design of commercial data processing (DP) systems. The third study stemmed from involvement on a KBS project being carried out by the Technology Division of the Trustees Saving Bank Group plc. The preliminary research highlighted poor human factors integration. In particular, there was a lack of early consideration of end-user requirements definition and user-centred evaluation. Instead concentration was given to the construction of the knowledge base and prototype evaluation with the expert(s). In response to this identified problem, a set of methods was developed that was aimed at encouraging developers to consider user interface requirements early on in a project. These methods were then applied in the two further projects, and their uptake within the overall development process was monitored. Experience from the two studies demonstrated that early consideration of user interface requirements was both feasible, and instructive for guiding future development work. In particular, it was shown a user interface prototype could be used as a basis for capturing requirements at the functional (task) level, and at the interface dialogue level. Extrapolating from this experience, a KBS life-cycle model is proposed which incorporates user interface design (and within that, user evaluation) as a largely parallel, rather than subsequent, activity to knowledge base construction. Further to this, there is a discussion of several key elements which can be seen as inhibiting the integration of human factors within KBS development. These elements stem from characteristics of present KBS development practice; from constraints within the commercial and industrial development environments; and from the state of existing human factors support.
Resumo:
Initially this thesis examines the various mechanisms by which technology is acquired within anodizing plants. In so doing the history of the evolution of anodizing technology is recorded, with particular reference to the growth of major markets and to the contribution of the marketing efforts of the aluminium industry. The business economics of various types of anodizing plants are analyzed. Consideration is also given to the impact of developments in anodizing technology on production economics and market growth. The economic costs associated with work rejected for process defects are considered. Recent changes in the industry have created conditions whereby information technology has a potentially important role to play in retaining existing knowledge. One such contribution is exemplified by the expert system which has been developed for the identification of anodizing process defects. Instead of using a "rule-based" expert system, a commercial neural networks program has been adapted for the task. The advantages of neural networks over 'rule-based' systems is that they are better suited to production problems, since the actual conditions prevailing when the defect was produced are often not known with certainty. In using the expert system, the user first identifies the process stage at which the defect probably occurred and is then directed to a file enabling the actual defects to be identified. After making this identification, the user can consult a database which gives a more detailed description of the defect, advises on remedial action and provides a bibliography of papers relating to the defect. The database uses a proprietary hypertext program, which also provides rapid cross-referencing to similar types of defect. Additionally, a graphics file can be accessed which (where appropriate) will display a graphic of the defect on screen. A total of 117 defects are included, together with 221 literature references, supplemented by 48 cross-reference hyperlinks. The main text of the thesis contains 179 literature references. (DX186565)