Ronald J. Brachman, Peter G. Selfridge, Loren G. Terveen, B. Altman, Alex Borgida, Fern Halper, Tom Kirk, A. Lazar, Deborah L. McGuinness, and Lori Alperin Resnick. Knowledge representation support for data archaeology. In First International Conference on Information and Knowledge Management, pages 457-464, Baltimore, MD, November 1992. 25  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... summarized by the KDD spectrum as follows: At one extreme, complete automation using machine learning to discover knowledge is advocated by one school of thought (e.g. data dredging) while at the other extreme, another school favors a dialectic and interactive orientation (e.g. data archaeology (Brachman et al. 1992, Brachman et al. 1993, Brachman et al. 1994) In between are machine assisted (e.g. Bhandari 1994) and human assisted knowledge discovery methods. This spectrum can be depicted in Figure 1. Data Dredging MachineAssisted Discovery Method HumanAssisted Discovery Method Data ....

....Visualization has been used in a limited extent in several KDD tools. The hope is that patterns and trends can be detected more easily than reading production rules. 1.2 Related Work Projects that have incorporated some form of visualization to aid in KDD include IMACS, MMV, and Netmap. IMACS ((Brachman et al. 1992), Brachman et al. 1993) Terveen 1993) Brachman et al. 1994) uses conventional graphs and plots as an interface for the analyst to segment data with mouse clicks, appearing as breaks in a graph to indicate segment boundaries. MVV (Mihalisin et al. 1991) uses bar charts (histogram within ....

Brachman, R.J., Selfridge, P.G., Terveen, L.G., Altman, B., Borgida, A., Halper, F., Kirk, T., Lazar, A., McGuinness, D.L., and Resnick, L.A. Nov 1992. Knowledge Representation Support for Data Archaeology. In Proceedings of the Intl. Conf. on Information & Knowledge Management (CIKM '92), Baltimore, MD, pp 457-464.

....that makes them especially suitable for reasoning about hierarchies of structured objects. DL systems have been used for building a variety of applications including systems supporting software management [20] browsing and querying of networked information sources [23] data archaeology [12], plan recognition [85] natural language understanding [8] and multimedia data description and classification [28] The grandfather of DL systems was KL ONE [11] Nowadays, a whole family of DL based knowledge representation systems has been built, like BACK [58] CLASSIC [10] KRIS [5] and LOOM ....

R. J. Brachman, P. G. Selfridge, L. G. Terveen, B. Altman, A. Borgida, F. Halpern, T. Kirk, A. Lazar, D. L. McGuinness, and L. A. Resnick. Knowledge representation support for data archaeology. In Y. Yesha, editor, Proceedings of the International Conference on Information and Knowledge Management CIKM-92, pages 457--464, 1992. 50

....We also present one view of safety of classic when it is used to query databases, and present a semantical basis for this view. 1 Introduction There is currently a great deal of interest in data mining applications, where a large source of data is combed for useful knowledge [15] Following [4], we distinguish between data mining approaches, where autonomous tools run freely over databases to search for interesting general properties and categories of the data, and the interactive data archaelogy approach, which (to quote the authors of [4] is an iterative, dialectic process that ....

.... combed for useful knowledge [15] Following [4] we distinguish between data mining approaches, where autonomous tools run freely over databases to search for interesting general properties and categories of the data, and the interactive data archaelogy approach, which (to quote the authors of [4]) is an iterative, dialectic process that involves constant human intervention . In the imacs [4] system, the authors use the example of marketeers searching a database of department store sales records for useful patterns. Likewise, public health officials may wish to explore a medical records ....

[Article contains additional citation context not shown here]

Brachman, R. J., Selfridge, P.G., Terveen, L.G., Altman, B., Borgida, A., Halper, F., Kirk, T., Lazar, A., McGuinness, D.L., and Resnick, L. A., Knowledge Representation Support for Data Archaeology, ISMM International Conference on Information and Knowledge Management, Baltimore, MD, November 1992.

No context found.

Ronald J. Brachman, Peter G. Selfridge, Loren G. Terveen, B. Altman, Alex Borgida, Fern Halper, Tom Kirk, A. Lazar, Deborah L. McGuinness, and Lori Alperin Resnick. Knowledge representation support for data archaeology. In First International Conference on Information and Knowledge Management, pages 457-464, Baltimore, MD, November 1992. 25

No context found.

Brachman, R. J., Selfridge, P. G., Terveen, L. G., Altman, B., Borgida, A., Halper, F., Kirk, T., Lazar, A., McGuinness, D. L., and Resnick, L. A., "Knowledge representation support for data archaeology," Proc. 1st Intl. Conf. on Information and Knowledge Management, Baltimore, MD, November, 1992, pp. 457--464.

No context found.

Ronald J. Brachman, Peter G. Selfridge, Loren G. Terveen, Boris Altman, Alex Borgida, Fern Halper, Thomas Kirk, Alan Lazar, Deborah L. McGuinness, Lori Alperin Resnick. "Knowledge Representation Support for Data Archaeology." In Proceedings of the First International Conference on Information and Knowledge Management, November, 1992.

....example, a database of customers and purchases, originally used for billing and inventory management, may now be seen as a resource for predicting customer behavior and defining customer sub groups, leading to more targeted, effective, and easily evaluated marketing campaigns. In a previous paper [7], we identified an approach to knowledge discovery in databases that we call data archaeology. The term emphasizes that this is a skilled process in which answers do not emerge in one pass, as fullblown nuggets, but rather evolve in an iterative, dialectic process that requires constant human ....

....one pass, as fullblown nuggets, but rather evolve in an iterative, dialectic process that requires constant human intervention. This process resembles the highly skilled work of an archaeologist. This paper describes a system called IMACS (Interactive Marketing Analysis and Classification System) [7] that addresses these problems, focusing on the user interface. While IMACS is a multi faceted system it uses knowledge representation technology to provide an expressive representation of data and to integrate data from multiple databases, employs novel techniques for translating from ....

[Article contains additional citation context not shown here]

Brachman, R.J., Selfridge, P.G., Terveen, L.G., Altman, B., Borgida, A., Halper, H., Kirk, T., Lazar, A., McGuinness, D.L., and Resnick, L.A. Knowledge Representation Support for Data Archaeology. In 1st Conference on Information and Knowledge Management. pp. 457-464. 1992.

.... (2) Merge multiple information sources, by describing their semantic content using DLs and thence deriving a common schema or a translation [CL93, LSK95] 3) Express queries so that the DML and the DDL are closely related, or to allow queries to be organized and checked for coherence [BGN89, Buc94a, Bra93]. 4) Query evaluation and optimization, through semantic query optimization [Bal95] finding and reusing the results of previous queries [BGN89, Buc94a] or just finding the relevant information sources to pass the query to [LSK95] However, with very few exceptions (notably [GL95] work until ....

R. Brachman, P. Selfridge, L. Terveen, B. Altman, A. Borgida, F. Halper, T. Kirk, A. Lazar, D. McGuinness and L. Resnick. Knowledge representation support for data archaeology, Int. J. of Intelligent and Cooperative Inf. Sys., 2(2), pp. 159--186, 1993.

....for tasks such as determining whether a concept is consistent, or whether it contains another one. There is growing evidence that DLs are useful in a variety of tasks dealing with data management 1 . As just one example, the Classic DL KBMS [9] was used in a prototype data exploration system [10] intended to support people in identifying interesting patterns in large accumulations of data from multiple sources; Classic was used, among others, as a language for capturing the domain model, as a unifying object centered representation of information This work was supported in part by NSF ....

R. Brachman, P.Selfridge, L.Terveen, B.Altman, A. Borgida, F. Halper, T.Kirk, A.Lazar, S.McGuiness, L.Resnick, "Knowledge Representation Support for Data Archaelogy", Int. J. of Intelligent and Cooperative Information Systems 2(2), June 1993, pp.159--186.

....solving, or user knowledge as in [16] or [15] In any case, the explanation is often presented as a chain of inferences by which a conclusion was reached. Description Logic based Systems (DLs) are another knowledge representation and reasoning technology that is increasingly used in applications [13, 10, 20, 7]. One benefit of DLs is that they have a richer representation formalism than standard rule based systems; at the same time, their reasoning mechanisms are both more extensive and more intricate. Moreover, for efficiency reasons, DLs are often implemented procedurally using all sorts of optimized ....

R. J. Brachman, P. G. Selfridge, L. G. Terveen, B. Altman, A. Borgida, F. Halper, T. Kirk, A. Lazar, D. L. McGuinness, and L. A. Resnick. Knowledge representation support for data archaeology. In First International Conference on Information and Knowledge Management, November 1992.

....many queries, possibly by teams of people, over an extended period of time. The DL based KBMS can automatically organize this large set of queries through the subsumption relationship, thereby allowing users to find identical or similar queries asked in the past, together with their answers (see [Brachman et al. 1992]) This is important if the queries may require a considerably long time to process. The operation of classifying a given new description with respect to some set of previously encountered descriptions is in fact standard in all DL reasoners, with various techniques for doing so surveyed in ....

....subset might vary from application to application. This approach requires a modular architecture for DL reasoners which is as easy to extend as, for example, a syntax directed translation scheme used in a programming language compiler. Such extensible architectures are discussed in [Borgida and Brachman 1992, Baader and Hanscke 1991] and the methodology of providing extensions is illustrated in [Borgida 1992b] Our conclusions in this section are that The conflicting desires between expressivenss and complexity of reasoning, although very real, need not be paralyzing: there is wide variety of ....

[Article contains additional citation context not shown here]

Brachman, R., P.Selfridge, L.Terveen, B.Altman, A. Borgida, F. Halper, T.Kirk, A.Lazar, S.McGuiness, L.Resnick, "Knowledge Representation Support for Data Archaelogy", submitted for publication.

....with the ease of use of form based methods, full support for iterative exploration, and means to reuse work and manage analyses over time. We have implemented a generic system called IMACS (Interactive Market Analysis and Classification System) that can be applied to different application domains [14]. In the remainder of this paper, we elaborate on the task of data archaeology, showing how current technology does not support it well; describe our approach and system, arguing that IMACS provides more adequate support; exemplify the capabilities of the system in a particular domain; and ....

Brachman, R. J., Selfridge, P. G., Terveen, L. G., Altman, B., Borgida, A., Halper, F., Kirk, T., Lazar, A., McGuinness, D. L., and Resnick, L. A., Knowledge Representation Support for Data Archaeology, Proceedings of the ISMM International Conference on Information and Knowledge Management (CIKM92) : 457-464, Baltimore, MD, November 1992.

....very many queries, possibly by teams of people, over an extended period of time. The DL based KBMS can automatically organize this large set of queries through the subsumption relationship, thereby allowing users to find identical or similar queries asked in the past, together with their answers [26]. This is important if the queries may require a considerably long time to process, or if users associate comments observations with queries. The operation of classifying a given new description with respect to some set of previously encountered descriptions is in fact standard in all ....

R. Brachman, P.Selfridge, L.Terveen, B.Altman, A. Borgida, F. Halper, T.Kirk, A.Lazar, S.McGuiness, L.Resnick, "Knowledge representation support for data archaelogy", Int. J. of Intelligent and Cooperative Information Systems 2(2), June 1993, pp.159--186.

No context found.

Ronald J. Brachman, et al. Knowledge representation support for data archaeology. In First International Conference on Information and Knowledge Management, pages 457--464, November 1992.

No context found.

Ronald J. Brachman, et al. Knowledge representation support for data archaeology. In First International Conference on Information and Knowledge Management, pages 457--464, November 1992.

No context found.

Ronald J. Brachman, et al. Knowledge representation support for data archaeology. In First International Conference on Information and Knowledge Management, pages 457--464, November 1992.

No context found.

Ronald J. Brachman, et al. Knowledge representation support for data archaeology. In First International Conference on Information and Knowledge Management, pages 457--464, November 1992.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC