Matheus et al, " Selecting and Reporting What is Interesting", In Fayyad et al. editors, Advances in Knowledge Discovery and Data Mining, AAAI Press, 1996.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....than the relatively simple rules that we use. Something can also be said of various KDD systems. Explora [14, 15] nds interesting instances of statistical patterns. The patterns discovered by 49er [16] are contingency tables, equations, and logical equivalences. The Key Finding Reporter (Ke r) [17, 18] tailored with a lot of domain knowledge, discovers and explains deviations, and gives recommendations for corrective actions. To the best of our knowledge, none ofthese systems isdirectly applicable todiscov ery in temporal sequences of events such as alarms. These systems also differ ....

C. J. Matheus, G. Piatetsky-Shapi ro, and D. McNeill, Selecting and reporting what is interest - ing. In U. M. Fayyad, G. Piatetsky - Shap iro, P. Smyth, and R. Uthurusamy (eds.), Advances in Knowledge Discovery and Data Mining , AAAI Press, Menlo Park, California pp. 495515, 1996.

....work on interestingness of knowledge tries to distinguish the potentially interesting patterns from others. There are a number of approaches studying the interestingness problem. Some researchers proposed approaches that determine interestingness of a discovered rule based on user s feedback [28, 21, 15], hence the measurement of interestingness is subjective. Some researchers developed various objective interestingness measures based on the statistics underlying the discovered patterns [25, 13, 12] With the quantitative objective interestingness measure, discovered patterns can be ranked and ....

C. J. Matheus, G. Piatetsky-Shapiro, and D. McNeill. Selecting and reporting what is interesting. In U. M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, editors, MIT Press, 1996.

....time. When knowledge discovery processes result in new, general concepts, these concepts can also be reflected in the ontology. In support of the data analysis phase, InfoSleuth provides generic analysis agents for performing data summarization [3] classification [9] and deviation detection [1, 29]. As illustrated in Figure 1, the execution of both data access and data analysis components is seamlessly controlled by the task planning and execution agent. 6.2 A Health care Application The InfoSleuth project is collaborating with the Health care Open Systems TriMs (HOST) consortium, ....

C. J. Matheus, G. Piatetsky-Shapiro, and D. McNeill, "Selecting and reporting what is interesting", In Usama M. Fayyad, Gregory Piatetsky-Shapiro, Padhraic Smyth and Ramasamy Uthurusamy (Editors) Advances in Knowledge Discovery and Data Mining, AAAI Press/The MIT Press Menlo Park, California, 1996.

....on interestingness of knowledge tries to distinguish the potentially interesting patterns from others. There are a number of approaches studying the interestingness problem. Some researchers proposed approaches that determine interestingness of a discovered rule based on the user s feedback [28] [21], 15] hence, the measurement of interestingness is subjective. Some researchers developed various objective interestingness measures based on the statistics underlying the discovered patterns [25] 13] 12] With the quantitative objective interestingness measure, discovered patterns can be ....

C.J. Matheus, G. Piatetsky-Shapiro, and D. McNeill, "Selecting and Reporting What Is Interesting," Advances in Knowledge Discovery and Data Mining, U.M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, eds., pp. 495-514, AAAI Press/The M1T Press, 1996.

....main concern is to apply some statistical measures over the rules. However, the statistical significance alone cannot guarantee interestingness of the rule. Rather it is the unexpectedness that can assure some flavor of interestingness. That is whywe find some works that employ deviation analysis [10, 11], one step towards unexpectedness to mine interesting rules. An objective measure is reliable in the sense that it avoids user s biased belief. However, it is not well accepted as it does not take into account user s interests or needs. Some improvementswould be made if the existing objective ....

C. J. Matheus, G. Piatetsky-Shapiro, and D. McNeill. Selecting and reporting what is interesting. In In Advances in Knowledge Discovery and Data Mining, pages 495--515. Menlo Park, Calif.: AAAI Press, 1996.

....the data. Thus, the additional steps such as data presentation, data selection, incorporating prior knowledge, and defining the semantics of the results obtained belong to KDD (cf. e.g. Fayyad et al. 14] Prominent examples of KDD applications in health care and finance include Matheus et al. [27] and Kloesgen [22] The importance of KDD research finds its explanation in the fact that the data collected in various fields such as biology, finance, retail, astronomy, medicine are extremely rapidly growing, while our ability to analyze those data has not kept up proportionally. KDD mainly ....

C.J. Matheus, G. Piatetsky-Shapiro, and D. McNeil, Selecting and reporting what is interesting, In (U.M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, Eds.), Advances in Knowledge Discovery and Data Mining pp. 495--515, Menlo Park, CA, AAAI Press, 1996.

....step, postprocessing of patterns can cause the user to look for some slightly modified types of patterns, etc. Efficient support for such iteration is one important development topic in KDD. Prominent applications of KDD include health care data, financial applications, and scientific data [39, 30]. One of the more spectacular applications is the SKICAT system [9] which operates on 3 terabytes of image data, producing a classification of approximately 2 billion sky objects into a few classes. The task is obviously impossible to do manually. Using example classifications provided by the ....

C. J. Matheus, G. Piatetsky-Shapiro, and D. McNeill. Selecting and reporting what is interesting. In U. M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, editors, Advances in Knowledge Discovery and Data Mining, pages 495 -- 515. AAAI Press, Menlo Park, CA, 1996.

....the structural dependence between variables. In a second, more detailed level, we have to find the quantitative dependency, that is, the strength of the dependency, regarding some scale. Probabilistic networks are the main representation for dependency modelling. ffl Trend and deviation detection [79]. This task consists in discovering from data about a system the most significant changes in its future evolution as well as the main trends that will guide it. Methods coming from statistics [4] signal processing, genetic algorithms, speech recognition, time series analysis [21] and dynamic ....

....have been made. Some important examples are: ffl Fraud detection [27, 76, 99] ffl Credit approval decision systems [41] ffl Investment analysis [25] ffl Marketing (Database Marketing) 5, 9, 43, 98] ffl Portfolio analysis [57] ffl Manufacturing process analysis [101] ffl Health care [79]. ffl Astronomy [37] ffl Geophysics [80] ffl Basketball scouting [10] ffl Spectral data analysis [15] There is a general set of criteria [40, 93] that can help in selecting whether KDD is applicable to a concrete project. In [40] they make a distinction between practical and technical ....

Christopher J. Matheus, Gregory Piatetsky-Shapiro, and Dwight McNeil. Selecting and Reporting What is Interesting. In Smyth Fayyad, Piatetsky-Shapiro and Uthurusamy, editors, Advances in Knowledge Discovery and Data Mining. MIT Press, 1996.

....and repeat the pattern discovery. The patterns discovered by 49er [27] are contingency tables, equations, and logical equivalencies. The user can interactively change the focus, e.g. independent and dependent variables, and require for a new pattern discovery. The Key Finding Reporter (Kefir) [17, 21] discovers and explains deviations, and gives recommendations for corrective actions. Applications of Kefir are tailored with a lot of domain knowledge to be aware of the interestingness criteria, corrective actions, etc. of the domain. Given a database from the domain, a Kefir based application ....

Christopher J. Matheus, Gregory Piatetsky-Shapiro, and Dwight McNeill. Selecting and reporting what is interesting. In Usama M. Fayyad, Gregory Piatetsky-Shapiro, Padhraic Smyth, and Ramasamy Uthurusamy, editors, Advances in Knowledge Discovery and Data Mining, pages 495 -- 515. AAAI Press, Menlo Park, CA, 1996.

....useful from the data. Thus, the additional steps such as data presentation, data selection, incorporating prior knowledge, and defining the semantics of the results obtained belong to KDD (cf. e.g. Fayyad et al. 1996b) Prominent examples of KDD applications in health care and finance include Matheus et al. 1996) and Kloesgen (1995) The importance of KDD research finds its explanation in the fact that the data collected in various fields such as biology, finance, retail, astronomy, medicine are extremely rapidly growing, while our ability to analyze those data has not kept up proportionally. KDD mainly ....

Matheus C.J., Piatetsky-Shapiro G., and McNeil D. 1996. Selecting and reporting what is interesting. In U.M. Fayyad, G. Piatetsky-Shapiro, P. Smyth and R. Uthurusamy, Eds., Advances in Knowledge Discovery and Data Mining, pp. 495--515, AAAI Press.

....and only a small fraction of which may be of interest to the users. Research work on interestingness of knowledge tried to distinguish the potentially interesting patterns from others. Some researchers proposed approaches that determine interestingness of a discovered rule based on user s feedback [16, 10, 14, 8], hence the measurement of interestingness is subjective. Researchers also developed various objective interestingness measures based on the statistics underlying the discovered patterns [12, 7] With the quantitative objective interestingness measure, discovered patterns can be ranked and less ....

C. J. Matheus, G. Piatetsky-Shapiro, and D. McNeill. Selecting and reporting what is interesting. In U. M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, editors, Advances in Knowledge Discovery and Data Mining, pages 495--514. AAAI Press / The MIT Press, 1996.

....step, postprocessing of patterns can cause the user to look for some slightly modified types of patterns, etc. Efficient support for such iteration is one important development topic in KDD. Prominent applications of KDD include health care data, financial applications, and scientific data [26, 19]. One of the more spectacular applications is the SKICAT system [7] which operates on 3 terabytes of image data, producing a classification of approximately 2 billion sky objects into a few classes. The task is obviously impossible to do manually. Using example classifications provided by the ....

C. J. Matheus, G. Piatetsky-Shapiro, and D. McNeill. Selecting and reporting what is interesting. In U. M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, editors, Advances in Knowledge Discovery and Data Mining, pages 495 -- 515. AAAI Press, Menlo Park, CA, 1996.

....the data. Thus, the additional steps such as data presentation, data selection, incorporating prior knowledge, and defining the semantics of the results obtained belong to KDD (cf. e.g. Fayyad et al. 14] Prominent examples of KDD applications in health care and finance include Matheus et al. [27] and Kloesgen [22] The importance of KDD research finds its explanation in the fact that the data collected in various fields such as biology, finance, retail, astronomy, medicine are extremely rapidly growing, while our ability to analyze those data has not kept up proportionally. KDD mainly ....

C.J. Matheus, G. Piatetsky-Shapiro, and D. McNeil, Selecting and reporting what is interesting, In (U.M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, Eds.), Advances in Knowledge Discovery and Data Mining pp. 495--515, Menlo Park, CA, AAAI Press, 1996.

....time. When knowledge discovery processes result in new, general concepts, these concepts can also be reflected in the ontology. In support of the data analysis phase, InfoSleuth provides generic analysis agents for performing data summarization [3] classification [9] and deviation detection [1, 29]. As illustrated in Figure 1, the execution of both data access and data analysis components is seamlessly controlled by the task planning and execution agent. 6.2 A Health care Application The InfoSleuth project is collaborating with the Health care Open Systems Trials (HOST) consortium, ....

C. J. Matheus, G. Piatetsky-Shapiro, and D. McNeill, "Selecting and reporting what is interesting", In Usama M. Fayyad, Gregory Piatetsky-Shapiro, Padhraic Smyth and Ramasamy Uthurusamy (Editors) Advances in Knowledge Discovery and Data Mining, AAAI Press/The MIT Press Menlo Park, California, 1996.

....iteration is also underlined. The iteration covers, however, either the whole process or at least the pattern discovery and presentation phases. To the best of our knowledge, none of the existing KDD systems supports the methodology proposed here. For instance, in Explora [9] 49er [16] and Kefir [11] the user can interactively change the focus, but that requires a new pattern discovery. Additionally, the approach of discovering all patterns can be contrasted with numerous methods, e.g. in machine learning, which aim directly at more focused discovery and produce one or at most few patterns ....

C. Matheus, G. Piatetsky-Shapiro, and D. McNeill. Selecting and reporting what is interesting. In Advances in Knowledge Discovery and Data Mining, pp. 495 -- 515. AAAI Press, Menlo Park, CA, 1996.

No context found.

Matheus et al, " Selecting and Reporting What is Interesting", In Fayyad et al. editors, Advances in Knowledge Discovery and Data Mining, AAAI Press, 1996.

No context found.

C.J. Matheus, G. Piatetsky-Shapiro, D. McNeill, Selecting and reporting what is interesting, in: U.M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, R. Uthurusamy (Eds.), Advances in Knowledge Discovery and Data Mining, AAAI Press, Menlo Park, CA, 1996, pp. 495.

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