The analysis of business data is often an ill-defined task characterized by large amounts of noisy data. Because of this, business data analysis must combine two kinds of intertwined tasks: exploration and analysis. Exploration is the process of finding the appropriate subset of data to analyze, and analysis is the process of measuring the data to provide the business answer. While there are many tools available both for exploration and for analysis, a single tool or set of tools may not provide full support for these intertwined tasks. We report here on a project that set out to understand a specific business data analysis problem and build an environment to support it. The results of this understanding are, first of all, a detailed list of requirements of this task; second, a set of capabilities that meet these requirements; and third, an implemented client-server solution that addresses many of these requirements and identifies others for future work. Our solution incorporates several novel perspectives on data analysis and combines a history mechanism with a graphical, re-usable representation of the analysis and exploration process. Our approach emphasizes using the database itself to represent as many of these functions as possible.

We identify an important type of software design knowledge that we call community-specific folklore and discuss problems with current approaches to managing it. We developed a general framework for a living design memory, built a design memory tool, and deployed the tool in a large software development organization. The tool effectively disseminates knowledge relevant to local software design practice. It is embedded in the organizational process to help ensure that its knowledge evolves as necessary. This work illustrates important lessons in building knowledge management systems, integrating novel technology into organizational practice, and carrying out research-development partnerships.

Corporate databases increasingly are being viewed as potentially rich sources of new and valuable knowledge. Various approaches to “discovering” or “mining” such knowledge have been proposed. Here we identify an important and previously ignored discovery task, which we call data archaeology. Data archaeology is a skilled human task, in which the knowledge sought depends on the goals of the analyst, cannot be specified in advance, and emerges only through an iterative process of data segmentation and analysis. We describe a system that supports the data archaeologist with a natural, object-oriented representation of an application domain, a powerful query language and database translation routines, and an easy-to-use and flexible user interface that supports interactive exploration. A formal knowledge representation system provides the core technology that facilitates database integration, querying, and the reuse of queries and query results.

In this panel session, the following topics are discussed: artificial intelligence in business; artificial intelligence in multimedia; neural networks as a tool for artificial intelligence: software engineering for knowledge-based systems: and artificial intelligence as a solution for software engineering.>

Computer vision algorithms are notorious for their computational expense. Distributed vision, the use of more than one processor, can decrease computation costs and speed up algorithms. There are various ways to do this, ranging from parallelism at the sensor level to true multiprocessor systems. This correspondence first describes a system of the latter type: a system of microprocessors on a high-speed bus. A canonical vision task, locating a number of objects and measuring certain two-dimensional features of those objects, serves as a benchmark test for the system. An algorithm for this task is presented. Performance measures are compared from implementations on the distributed system, a Vax 11/750, and a Vax 11/780. Results indicate that three microprocessors outperform a Vax 11/780 at this task. Finally, other more interesting distributed algorithms are briefly discussed.

Two approaches to organ detection in ab dominal computerized tomography scans, one local and one global, have been developed. The first involves a boundary-delineating algorithm which operates homogeneously on entire images and coordinates the use of multiple local criteria for advanced edge detection. The second involves an iterative, adaptive boundary-delineating algorithm suitable for single organ detection and amenable to the incorporation of organ-specific knowledge. Regional results of the first (global) algorithm can initialize the second (local) algorithm. Algorithms were tested on mathematical phantoms prior to application on clinical patient data.