Automated spoken dialog systems require systematic procedures for evaluating performance and diagnosing problems. We present an interactive tool that provides graphical views of how callers navigate through such syste...
详细信息
Automated spoken dialog systems require systematic procedures for evaluating performance and diagnosing problems. We present an interactive tool that provides graphical views of how callers navigate through such systems, enabling fine-grained analysis for system evaluation and business intelligence. The input is a feed of call-logs. The output is an empirical dialog trajectory analysis represented as stochastic finite state machines, accessible via the web. Complexity is managed by an automatic tokenization procedure that hides fine details until needed. Users can generate selective views of parts of the dialog at high resolution (with access to call data), or zoom out to a summary. The tool provides dialog system developers with all the information they need from a single source, and is in use with directed-dialog and natural-language applications.
The development of tools that can increase the productivity of computational chemists is of paramount importance to the pharmaceutical industry. Reducing the cost of drug research benefits consumer and company alike. ...
详细信息
ISBN:
(纸本)3540260323
The development of tools that can increase the productivity of computational chemists is of paramount importance to the pharmaceutical industry. Reducing the cost of drug research benefits consumer and company alike. Apt Apprentice is a visual programming paradigm designed to reduce the overhead associated with creating software to implement algorithms in the dataanalysis phase of rational drug design. It draws on both standard programming language environments and programming by demonstration. The approach of Apt Apprentice and an example of its use in implementing a linear algebra routine are described.
The Federal Aviation Administration (FAA) air traffic control system uses Decision Support Tools (DSTs) such as the User Request Evaluation Tool (URET) and the Traffic Management Advisor (TMA) to assist air traffic co...
详细信息
The Federal Aviation Administration (FAA) air traffic control system uses Decision Support Tools (DSTs) such as the User Request Evaluation Tool (URET) and the Traffic Management Advisor (TMA) to assist air traffic controllers to separate air traffic. DSTs predict aircraft flight paths and foretell potential conflicts. The accuracy of the trajectories generated by the DSTs determines their overall performance. The Conflict Probe Assessment Team, located at the FAA William J. Hughes Technical Center in New Jersey, has measured the accuracy of the TMA and URET trajectory modelers. And they will continue to study the accuracy of trajectory modelers in planned systems. A methodology has been developed to measure the accuracy of trajectories. Various algorithms and relational databases have been fashioned to investigate and calculate the performance accuracy of trajectory prediction. This paper presents a new graphical tool called the Trajectory Graphical User Interface (TrajectoryGUI), which was jointly developed by the Rowan University's Software Engineering, Graphics, and visualization Research Group and the FAA, to assist the investigators in measuring the accuracy. The tool provides assistance through an easy to use graphical user interface. The key features of the tool are its ability to graphically plot from a plethora of different flights and trajectories, view the plotted path, both in horizontal and vertical dimensions, zoom in and out, apply time tags for comparisons, and export plots and statistical analysis results of the investigation for external presentation. The development of this tool has resulted in a thorough process of evaluating the trajectory prediction accuracy of DSTs, It supplements the existing dataanalysis environment to a visualizationanalysis environment. An analyst can use TrajectoryGUI to package the trajectory accuracy results and illustrate the inaccuracy.
In this paper, we develop a new active contour model for image selective segmentation. The model adopts cascade anisotropic diffusion preprocessing and a selective term in level set function. Cascade anisotropic diffu...
详细信息
ISBN:
(纸本)0769523927
In this paper, we develop a new active contour model for image selective segmentation. The model adopts cascade anisotropic diffusion preprocessing and a selective term in level set function. Cascade anisotropic diffusion filtering is powerful and flexible to enhance image for various segmentation tasks. The selective term in level set function can evolve a single curve to capture a selective segmentation region which we are interested in. This is useful for intentional segmentation tasks. We can also realize the multi-region segmentation by varying selecting term conditions. Furthermore, we saliently speed the process of the new algorithm by using AOS scheme in cascade anisotropic diffusion filtering and discarding mean curvature motion in level set function. We illustrate the performance of our segmentation method on images generated by different modalities.
A Comparison method for Web pages in terms of visual similarity is proposed Conventional Web Information retrieval/gathering systems, such as search engines, extract keywords from HTML source files, based on which the...
详细信息
ISBN:
(纸本)076952415X
A Comparison method for Web pages in terms of visual similarity is proposed Conventional Web Information retrieval/gathering systems, such as search engines, extract keywords from HTML source files, based on which the similarity between pages is calculated. The extracted keywords are considered as semantic features representing the contents of Web pages. On the other hand, visual feature of Web pages is as important as semantic feature, because HTML is designed for visualizing a Web page in understandable manner for humans. The proposed method compares the layouts of Web pages based on image processing and graph matching. The experimental results show that the accuracy of layout analysis is 91.6% in average, and the visual similarity calculated by the proposed method is closer to the visual judgment by test subjects than color-based comparison method.
Information visualisation applications have been facing ever-increasing demands as the amount of available information has increased exponentially. With this, the number and complexity of visualisation tools for analy...
详细信息
ISBN:
(纸本)1595930361
Information visualisation applications have been facing ever-increasing demands as the amount of available information has increased exponentially. With this, the number and complexity of visualisation tools for analysing and exploring data has also increased dramatically, making development and evolution of these systems difficult. We describe an investigation into reusing technology developed for computer games to create collaborative information visualisation tools. A framework for using game engines for information visualisation is presented together with an analysis of how the capabilities and constraints of a game engine influence the mapping of data into graphical representations and the interaction with it. Based on this research a source code comprehension tool was implemented using the Quake 3 computer game engine. It was found that game engines can be a good basis for an information visualisation tool, provided that the visualisations and interactions required meet certain criteria, mainly that the visualisation can be represented in terms of a limited number of discrete, interactive, and physical entities placed in a static 3-dimensional world of limited size. Copyright 2005 ACM.
In the first part of this article, Communities of Practice are conceptually positioned as a very important and successful element of corporate Knowledge Management. By utilizing IT platforms they enable a direct conne...
详细信息
In the first part of this article, Communities of Practice are conceptually positioned as a very important and successful element of corporate Knowledge Management. By utilizing IT platforms they enable a direct connection of knowledge workers and the transfer and reuse of tacit expertise to geographically remote business problems. Although current Community Software provides its members with many sophisticated features, facilitators or moderators still lack functionalities to monitor, evaluate and communicate the development of their expert networks. After discussing the requirements of this special target group, this contribution concentrates on electronic discussions and proposes a software system for automatically analyzing the structure and value of Knowledge Communities by extracting available electronic data about their communication network. This includes the entities employee, topic, and document and their many relationships. Insightful structural visualizations based on theories of Network analysis are introduced. They can be accessed and manipulated in a Management Cockpit to improve the transparency in communities.
The cyclogram of human gait uncovers the relationship between thigh and knee motion. Because it is believed to be invariant for the walking pattern of the same person it leads to a better understanding of gait. If we ...
详细信息
The cyclogram of human gait uncovers the relationship between thigh and knee motion. Because it is believed to be invariant for the walking pattern of the same person it leads to a better understanding of gait. If we use it as the signature in gait recognition and verification, it could lead to an automatic person recognition system using video footage from security cameras. The body contours are extracted by image subtraction and edge detection and rotation angles of thigh and lower leg are derived by regression analysis. Then we apply a weighted moving average and a low-pass filter to smooth the angle data. Cyclograms are generated as the thigh angle vs. knee angle and are fitted by non-uniform B-Spline curves. To compare the signatures between two gaits, the differences of shape and phase of the cyclogram are calculated using the point projection method and extreme points of curves. Finally, classification is done via the k-nearest classifier and cross-validation with the leave-one-out rule.
The proceedings contain 59 papers. The topics discussed include: learning from animated diagrams: learners opinions of cal materials and environments;evaluation of a learning repository system approach established in ...
详细信息
ISBN:
(纸本)0889864896
The proceedings contain 59 papers. The topics discussed include: learning from animated diagrams: learners opinions of cal materials and environments;evaluation of a learning repository system approach established in the schools and faculties of information technology and computer science in three large universities;ubiquitous data sampling and augmented real-time monitoring system;a visual communication design curriculum for the digital age;web-based lecturing: a case study of the CUSB certificate program in translation;an algorithm for extracting important sentences from papers submitted by students;improving the biomedical engineering design course process;technology, mathematics, and science: what do teachers and students think?;a simple computer model based on visualization, analogy and collaboration;an analysis of the design features of three mixed-mode courses in a master's degree programme;and experimental study on the reuse of learning objects and teaching practices.
This paper describes a novel progressive and hierarchical representation for densely built urban areas. It aims at navigating through huge urban environments with network-based client-server systems. Thanks to this me...
详细信息
This paper describes a novel progressive and hierarchical representation for densely built urban areas. It aims at navigating through huge urban environments with network-based client-server systems. Thanks to this method, navigation is no longer limited to walkthrough, but allows flying over the city. This can be achieved through a set of dedicated algorithms allowing the decomposition of city into a multi resolution representation. The method efficiently exploits the fact that automated modelling technologies of urban scenes are generally based on the processing of 2D data (video, photographs, cadastral map,...), to model 2D1/2 data (building footprint, height, altitude,...) which turns out to be more compact. From this pre-computed representation of buildings, the server progressively sends the perceptible details to the client for all the regions visible from a given viewpoint.
暂无评论