Predicting blood concentration levels of pharmaceutical agents in human subjects can be made difficult by missing data and variability within and between human subjects. Biometricians use a variety of software tools to analyze pharmacokinetic information in order to conduct research about a pharmaceutical agent. This paper is the comparison between using a feedforward backpropagation neural network to predict blood serum concentration levels of the drug tobramycin in pediatric cystic fibrosis and hemotologicâ oncologic disorder patients with the most commonly used software for analysis of pharmacokinetics, NONMEM. Mean squared standard error is used to establish the comparability of the two estimation methods. The motivation for this research is the desire to provide clinicians and pharmaceutical researchers a cost effective, user friendly, and timely analysis tool for effectively predicting blood concentration ranges in human subjects.

The problems of information overload and vocabulary differences have become more pressing with the emergence of increasingly popular Internet services. The main information retrieval mechanisms provided by the prevailing Internet WWW software are based on either keyword search (e.g., the Lycos server at CMU, the Yahoo server at Stanford) or hypertext browsing (e.g., Mosaic and Netscape). This research aims to provide an alternative concept-based categorization and search capability for WWW servers based on selected machine learning algorithms. Our proposed approach, which is grounded on automatic textual analysis of Internet documents (homepages), attempts to address the Internet search problem by first categorizing the content of Internet documents. We report results of our recent testing of a multilayered neural network clustering algorithm employing the Kohonen self-organizing feature map to categorize (classify) Internet homepages according to their content. The category hierarchies created could serve to partition the vast Internet services into subject-specific categories and databases and improve Internet keyword searching and/or browsing.

Areas related to nanotechnology, or nanoscale science and engineering (NSSE), have experienced tremendous growth over the past few years. While there are a large variety of useful resources available on the Web, such information are usually distributed and difficult to locate, resulting in the problem of information overload. To address the problem, we developed the NanoPort system, an integrated Web portal aiming to provide a one-stop shopping service to satisfy the information needs of researchers and practitioners in the field of NSSE [1]. We believe that the approaches taken also can be applied to other domains.

As in many other applications, the rapid proliferation and unrestricted Web-based publishing of health-related content have made finding pertinent and useful healthcare information increasingly difficult. Although the development of healthcare information retrieval systems such as medical search engines and peer-reviewed medical Web directories has helped alleviate this information and cognitive overload problem, the effectiveness of these systems has been limited by low search precision, poor presentation of search results, and the required user search effort. To address these challenges, we have developed a domain-specific meta-search tool called Cancer Spider. By leveraging post-retrieval document clustering techniques, this system aids users in querying multiple medical data sources to gain an overview of the retrieved documents and locating answers of high quality to a wide spectrum of health questions. The system presents the retrieved documents to users in two different views: (1) Web pages organized by a list of key phrases, and (2) Web pages clustered into regions discussing different topics on a two-dimensional map (self-organizing map). In this paper, we present the major components of the Cancer Spider system and a user evaluation study designed to evaluate the effectiveness and efficiency of our approach. Initial results comparing Cancer Spider with NLM Gateway, a premium medical search site, have shown that they achieved comparable performances measured by precision, recall, and F-measure. Cancer Spider required less user searching time, fewer documents that need to be browsed, and less user effort.

Law enforcement agencies across the United States have begun to focus on innovative knowledge management technologies to aid in the analysis of criminal information. The use of such technologies can serve as intelligence tools to combat criminal activity by aiding in case investigation or even by predicting criminal activity. Funded by the National Institute of Justice, the University of Arizonaâ s Artificial Intelligence Lab has teamed with the Tucson Police Department (TPD) to develop the Coplink Concept Space application, which serves to uncover relationships between different types of information currently existing in TPDâ s records management system. A small-scale field study involving real law enforcement personnel indicates that the use of Coplink Concept Space can reduce the time spent on the investigative task of linking criminal information as well as provide strong arguments for expanded development of similar knowledge management systems in support of law enforcement.

This paper describes a content-based image retrieval digital library that supports geographical image retrieval over a testbed of 800 aerial photographs, each 25 megabytes in size. In addition, this paper also introduces a methodology to evaluate the performance of the algorithms in the prototype system. The major contributions of this paper are two. 1) We suggest an approach that incorporates various image processing techniques including Gabor filters, image enhancement, and image compression, as well as information analysis technique such as self-organizing map (SOM) into an effective large-scale geographical image retrieval system. 2) We present two experiments that evaluate the performance of the Gaborfilter- extracted features along with the corresponding similarity measure against that of human perception, addressing the lack of studies in assessing the consistency between an image representation algorithm or an image categorization method and human mental model.

A major problem that decision makers are facing in an information-rich society is how to absorb, filter and make effective use of available data. The problem caused by information overflow could lead to the losses of competitiveness. This paper presents a knowledge-based approach to building an issues identifier to help investors overcome information overflow problems when dealing with very large on-line financial databases. The proposed software system is able to extract critical issues from the on-line financial databases. The system was developed based on a number of techniques: automatic indexing, concept space genemtion, and neural network classification. In this paper, we describe how these techniques are used to extract subject descriptors, their semantic relationships, and the related texts (documents or paragraphs) to each descriptor. The proposed system has been tested with the annual reports from thirteen of the largest international banks.

Recommender systems are being widely applied in many application settings to suggest products, services, and information items to potential consumers. Collaborative filtering, the most successful recommendation approach, makes recommendations based on past transactions and feedback from consumers sharing similar interests. A major problem limiting the usefulness of collaborative filtering is the sparsity problem, which refers to a situation in which transactional or feedback data is sparse and insufficient to identify similarities in consumer interests. In this article, we propose to deal with this sparsity problem by applying an associative retrieval framework and related spreading activation algorithms to explore transitive associations among consumers through their past transactions and feedback. Such transitive associations are a valuable source of information to help infer consumer interests and can be explored to deal with the sparsity problem. To evaluate the effectiveness of our approach, we have conducted an experimental study using a data set from an online bookstore. We experimented with three spreading activation algorithms including a constrained Leaky Capacitor algorithm, a branch-and-bound serial symbolic search algorithm, and a Hopfield net parallel relaxation search algorithm. These algorithms were compared with several collaborative filtering approaches that do not consider the transitive associations: a simple graph search approach, two variations of the user-based approach, and an item-based approach. Our experimental results indicate that spreading activation-based approaches significantly outperformed the other collaborative filtering methods as measured by recommendation precision, recall, the F-measure, and the rank score.We also observed the over-activation effect of the spreading activation approach, that is, incorporating transitive associations with past transactional data that is not sparse may “dilute” the data used to infer user preferences and lead to degradation in recommendation performance.

A prototype tool classifies output from an electronic meeting system into a manageable list of concepts, topics, or issues that a group can further evaluate. In an experiment with output from GroupSystems electronic meeting system, the tool's recall ability was comparable to that of a human facilitator, but took roughly a sixth of the time.

This paper discusses several data mining algorithms and techniques that we have developed at the University of Arizona Artificial Intelligence Lab.We have implemented these algorithms and techniques into several prototypes, one of which focuses on medical information developed in cooperation with the National Cancer Institute (NCI) and the University of Illinois at Urbana-Champaign.We propose an architecture for medical knowledge information systems that will permit data mining across several medical information sources and discuss a suite of data mining tools that we are developing to assist NCI in improving public access to and use of their existing vast cancer information collections.