Ronald J. Brachman, Peter G. Selfridge, Loren G. Terveen, Boris Altman, Alex Borgida, Fern Halpern, Thomas Kirk, Alan Lazar, Deborah L. McGuinness, and Lori Alperin Resnick. Integrated support for data archaeology. In ISMM International Conference on Information and Knowledge Management, Baltimore, MA, November 1992.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... 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 Archaelogy Figure 1: The KDD Spectrum In practice, ....

....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 histogram within histogram) and slider bars (with horizontal scales) ....

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. Aug 1994. Integrated Support for Data Archaeology. In Proceedings of the Proc. AAAI-93 Knowledge Discovery in Databases Wkshp, Seattle, WA.

.... 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 Archaelogy Figure 1: The KDD ....

....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 histogram within histogram) ....

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. 1993. Integrated Support for Data Archaeology. Intl. Journal of Intelligent & Cooperative Information Systems.

....them with other modules as part of a complete application. Generic, multitask tools perform a variety of discovery tasks, typically combining classification (perhaps using more than one approach) visualization, query retrieval, clustering, and more. Examples include Clementine [7] IMACS [4], MLC , MOBAL, and Recon. These tools support more of the KDD process and simplify embedding of discovered knowledge into an application the business user can use. The target user of such tools is usually a sophisticated analyst who understands data manipulation. While these tools typically ....

Brachman, R., Selfridge, P., Terveen, L., Altman, B., Borgida, A., Halper, F., Kirk, T., Lazar, A., McGuinness, D., and Resnick, L. Integrated support for data archaeology. Int. J. Intell. Coop. Inf. Syst. 2, 2 (June 1993), 159--185.

....of the progress of the query and guiding toward the possible actions which can be performed. This can be separated into two levels. At the data level, the emphasis is on feedback concerning the answering of the query. Such an approach has been used with traditional databases [9] The IMACS project [7] used a CLASSIC knowledge base to support data mining and knowledge discovery, providing more sophisticated feedback. Alternatively, we can provide feedback at a meta or schema level, constraining and guiding the user based on knowledge about the information model for example offering suitable ....

R. Brachman, P. Selfridge, L. Terveen, B. Altman, A. Borgida, F. Halper, T. Kirk, A. Lazar, D. McGuinness, and L. Renick. Integrated Support for Data Archaeology. International Journal of Applied and Cooperative Information Systems, 2(2):159--185, 1993.

....a fair amount of hype but as yet al..most no practitioners. I suspect this has happened because people assume TDM is a natural extension of the slightly less nascent field of data mining (DM) also known as knowledge discovery in databases (Fayyad and Uthurusamy, 1999) and information archeology (Brachman et al. 1993). Additionally, there are some disagreements about what actually constitutes data mining. It turns out that mining is not a very good metaphor for what people in the field actually do. Mining implies extracting precious nuggets of ore from otherwise worthless rock. If data mining really followed ....

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. 1993. Integrated support for data archaeology.

....to organise queries automatically and views, hence supporting data exploration and query optimisation. Description Logics have been used in a wide variety of applications including the representation of complex schemas for cars [30] software management [8] and medicine [27] data archaeology [6], mediation between heterogeneous data stores [2] and database querying [16] However, interacting with Description Logic implementations can often prove difficult. In the past, solutions involved embedding the logic in large monolithic systems with all the attached problems of maintenance and ....

R. Brachman, P. Selfridge, L. Terveen, B. Altman, A. Borgida, F. Halper, T. Kirk, A. Lazar, D. McGuinness, and L. Renick. Integrated Support for Data Archaeology. International Journal of Applied and Cooperative Information Systems, 2(2):159--185, 1993.

....Query sliders have been added. the class of queries can be anticipated. To complete the analogy with VQE, Butterfly would be able to modify queries by operations on query results. For instance, additional keywords associated with a citation might be dragged into the query. 5. 5 IMACS IMACS [4] has very similar goals to Visage VQE. It supports analysts iterative exploration and visualization of data, including query reuse. It s foundation is the knowledge representation system CLASSIC, which is much more sophisticated than the object oriented database we use. The primary advantage of ....

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. Integrated support for data archaeology. International Journal of Intelligent and Cooperative Information Systems, 2(2):159-185, 1993.

....(middle ground) type of discovery that is described in the middle column of Figure 1. This type of user involvement provides some information to the discovery system of what to look for so that the search engine can limit the search space using this information. It has been argued in [Bra93] that knowledge discovery (referred to as data archaeology in [Bra93] is a human centered task that cannot be completely automated. In addition, Bra93] argues that the manual approach of using SQL queries is not an appropriate 1 Type of Discovery Manual Semi Automatic Automatic (by the ....

....middle column of Figure 1. This type of user involvement provides some information to the discovery system of what to look for so that the search engine can limit the search space using this information. It has been argued in [Bra93] that knowledge discovery (referred to as data archaeology in [Bra93] is a human centered task that cannot be completely automated. In addition, Bra93] argues that the manual approach of using SQL queries is not an appropriate 1 Type of Discovery Manual Semi Automatic Automatic (by the user) middle ground) by the KDD system) User Involvement: Discovery ....

[Article contains additional citation context not shown here]

Brachman, R. J. et al. Integrated support for data archaeology. International Journal of Intelligent and Cooperative Information Systems, 2(2):159--185, 1993. 5

....from the user, largely because the class of queries can be anticipated. To complete the analogy with VQE, Butterfly would be able to modify queries by operations on query results. For instance, additional keywords associated with a citation might be dragged into the query. 5. 5 IMACS IMACS [4] has very similar goals to Visage VQE. It supports analysts iterative exploration and visualization of data, including query reuse. It s foundation is the knowledge representation system CLASSIC, which is much more sophisticated than the object oriented database we use. The primary advantage of ....

Ronald J Brachman, Peter G Selfridge, Loren G Terveen, Boris Altman, Alex Borgida, Fern Halper, Thomas Kirk, Alan Lazar, Deborah L McGuinness, and Lori Alperin Resnick. Integrated support for data archaeology. International Journal of Intelligent and Cooperative Information Systems, 2(2):159--185, 1993.

.... in ALC R is Pspace complete complete [Sattler, 1996] and determining subsumption in propositional dynamic logic is exponential time complete [Pratt, 1979] This and related complexity problems have lead some developers of description logic systems to use less expressive description logics [Brachman et al. 1993]. However, it is possible to build practical description logic systems based on expressive description logics [Baader and Hollunder, 1991b; Bresciani et al. 1995; Horrocks, 1997] that have this sort of computationally intractable subsumption problem. Systems that are based on description logics ....

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. Renick. Integrated support for data archaeology. International Journal of Applied and Cooperative Information Systems, 2(2):159--185, 1993.

....records and provide methods for querying them to obtain all records whose content satisfies the user s query. More recently however, researchers in Knowledge Discovery in Databases (KDD) have provided a new family of tools for accessing information in databases (e.g. Anand and Khan, 1993; Brachman et al., 1993; Frawley et al., 1991; Kl sgen, 1992) The goal of such work, often called data mining, has been defined as the nontrivial extraction of implicit, previously unknown, and potentially useful information from given data (PiatetskyShapiro and Frawley, 1991) Work in this area includes applying ....

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., 1993. Integrated Support for Data Archaeology. International Journal of Intelligent and Cooperative Information Systems 2(2):159-185.

....in the form of structured records and provide methods for querying them to obtain all records whose content satisfies the user s query. More recently however, researchers in Knowledge Discovery in Databases (KDD) have provided a new family of tools for accessing information in databases [1,3,15,19,20]. The goal of such work, often called data mining, has been defined as the nontrivial extraction of implicit, previously unknown, and potentially useful information from given data [15] Work in this area includes applying machinelearning and statistical analysis techniques towards the ....

Brachman, R. J.; Selfridge, P.G.; Terveen, L.G.; Altman, B.; Borgida, A.; Halper, F.; Kirk, T.; Lazar, A.; McGuinness, D.L.; Resnick, L.A.: Integrated Support for Data Archaeology. International Journal of Intelligent and Cooperative Information Systems, (1993)2(2):159185.

.... axioms results in EXPTIME complete subsumption 4 [38, 48, 29] Determining subsumption in propositional dynamic logic is also EXP TIME complete [39] These and related complexity problems have lead some developers of description logic systems to use less expressive description logics [9]. However, although the theoretical complexity results are discouraging, empirical analyses of real applications have shown that the kinds of construct that lead to worst case intractability rarely occur in practice [35, 25, 46, 28] and it has proved possible to build practical description logic ....

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. Renick. Integrated support for data archaeology. International Journal of Applied and Cooperative Information Systems, 2(2):159--185, 1993.

....front end can be more adequate than usual KB interfaces. ffl Data Archaeology. In data archaeology the main task is the search and extraction of previously ignored knowledge from several and possibly large databases, by means of an interactive and iterative process of analysis and refinement [Brachman et al. 1992]. In this case, the use of intelligent front ends based on DL can be helpful [Devanbu,1993] The use of friendly and powerful interfaces is even more important when data are widely and easily available to several kinds of users, but scarcely inter organized and spread over several sites as in the ....

Ronald J. Brachman, Peter G. Selfridge, Loren G. Terveen, Boris Altman, Alex Borgida, Fern Halpern, Thomas Kirk, Alan Lazar, Deborah L. McGuinness, and Lori Alperin Resnick. Integrated support for data archaeology. In ISMM International Conference on Information and Knowledge Management, Baltimore, MA, November 1992.

....of interesting patterns, visualization of results, etc. 4. Putting the discovered knowledge into use. The TASA system incorporates components for two parts of the KDD process: pattern discovery (or data mining) and presentation of the knowledge. KDD is an iterative and interactive process [3, 4]. No realistic KDD system can be expected to discover useful knowledge without interaction with the user: knowing the interests and using the background knowledge of users is vital for succesful knowledge discovery. Also, iteration is essential: after receiving some knowledge the user is better ....

Ronald J. Brachman, Peter G. Selfridge, Loren G. Terveen, Boris Altman, Alex Borgida, Fern Halper, Thomas Kirk, Alan Lazar, Doborah L. McGuinness, and Lori Alperin Resnick. Integrated support for data archaeology. In Knowledge Discovery in Databases, Papers from the 1993 AAAI Workshop (KDD'93), pages 197 -- 211, Washington, D.C., July 1993.

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. Integrated Support for Data Archaeology." In International Journal of Intelligent and Cooperative Information Systems, 2:2 1993, pages 159--185.

No context found.

Ronald J. Brachman, Peter G. Selfridge, Loren G. Terveen, Boris Altman, Alex Borgida, Fern Halpern, Thomas Kirk, Alan Lazar, Deborah L. McGuinness, and Lori Alperin Resnick. Integrated support for data archaeology. In ISMM International Conference on Information and Knowledge Management, Baltimore, MA, November 1992.

No context found.

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. Renick. Integrated support for data archaeology. International Journal of Applied and Cooperative Information Systems, 2, 159--185, 1993.

No context found.

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. Renick. Integrated support for data archaeology. International Journal of Applied and Cooperative Information Systems, 2(2):159--185, 1993.

No context found.

Ronald 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, Integrated Support for Data Archaeology, Intl. J. of Intelligent & Cooperative Information Systems, Vol. 2, No. 2, pp. 159 - 185.

No context found.

Ronald J. Brachman, et al. Integrated support for data archaeology. International Journal of Intelligent and Cooperative Information Systems, 2:159--185, 1993.

No context found.

Ronald J. Brachman, et al. Integrated support for data archaeology. International Journal of Intelligent and Cooperative Information Systems, 2:159--185, 1993.

No context found.

Brachman, R.J., Selfridge, L., Terven, L., Altman, B., Halper, F., Kirk, T., Lazar, A., McGuiness, D., & Resnick, L. (1993). Integrated Support for Data Archaeology. Proceedings 1993 AAAI Workshop on Knowledge Discovery in Databases, pp. 197-212.

No context found.

R. J. Brachman, P. G. Selfridge, L. G. Terveen, B. Altman, A. Borgida, F. Halper, T. Kirk, A. Lazar, D. L. McGuiness, and L. A. Resnick. Integrated support for data archaeology. In Proc. of AAAI-94 Knowledge Discovery in Databases Workshop, Seattle, WA, 1994.

No context found.

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. Renick. Integrated Support for Data Archaeology. International Journal of Applied and Cooperative Information Systems, 2(2):159--185, 1993.

First 50 documents

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