J.W. Grzymala--Busse, On the Unknown Attribute Values in Learning from Examples, In Proceedings of Sixth International Symposium Methodologies for Intelligent Systems, 368--377, October 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....noise. 5.2.3.3 Experiments with Increasing Level of Missing Values Most of the real world datasets contain missing (unknown) feature values and the percentage of missing values are shown in Table A.1. In order to cope with instances that contain missing values, several methods have been proposed [30, 55, 56, 57, 58]. These methods can be summarized as: ffl Ignoring instances which have unknown feature values. ffl Assuming an additional special value for unknown attribute values. ffl Using probability theory by utilizing information provided by context. ffl Generating additional instances for all possible ....

.... highest vote, received by Feature values of test instance 1: F[1] 2 F[2] 3 F[3] 3 F[4] 3 F[5] 3 F[6] 0 F[7] 0 F[8] 0 F[9] 3 F[10] 3 F[11] 0 F[12] 0 F[13] 0 F[14] 0 F[15] 0 F[16] 0 F[17] 3 F[18] 2 F[19] 2 F[20] 3 F[21] 3 F[22] 3 F[23] 1 F[24] 3 F[25] 0 F[26] 0 F[27] 0 F[28] 0 F[29] 0 F[30]:0 F[31] 0 F[32] 1 F[33] 0 F[34] 34 Classes: 1] 2] 3] 4] 5] 6] Votes of Feature[1] 0.16 0.16 0.17 0.18 0.15 0.18 Votes of Feature[2] 0.36 0.27 0.15 0.07( 0.05( 0.10 Votes of Feature[3] 0.34 0.05( 0.38 ....

[Article contains additional citation context not shown here]

J.W. Grzymala--Busse, On the Unknown Attribute Values in Learning from Examples, In Proceedings of Sixth International Symposium Methodologies for Intelligent Systems, 368--377, October 1991.

....for each specialized application. A better alternative of designing an expert system would be to construct and inductive algorithm that can, from a carefully chosen sample of expert decisions, infer and refine decision rules automatically independent of the subject of interest. The papers [5] 6][7][10] describe some of the more recent research efforts made in this area. Quinlan[1] suggested an inductive algorithm based on the statistical theory of information originally proposed by Shannon. The entropy function is used as a measure of uncertainty in the classification of objects ....

J. W. Grzymala-Busse, et al., On the Unknown Attribute Values in Learning from Examples, Proc. Int. Symp. on Methodologies for Intelligent Systems or Lecture Notes in Artificial Intelligence, Z. W. Ras, et al.(eds.), vol. 542, pp. 368-377, 1991

....it may not be computationally feasible to find a reduct. Furthermore, finding just a single reduct may be too restrictive for some data analysis problems. One plausible approach is to utilize the idea of reduct defined in the previous subsection[6] To handle missing values, Grzymala Busse [8] has transformed a given decision table with unknown values to a new and possibly inconsistent decision table by replacing the unknown attribute value with all possible values of that attribute. In other words, he reduced the missing value problem to that of learning from inconsistent examples. He ....

GRZYMALA-BUSSE, J. W. On the unknown attribute values in learning from examples. In Proceedings of Methodologies for Intelligent Systems, Z. W. Ras and M. Zemankowa, Eds., Lecture Notes in AI, 542. Springer-Verlag, New York, 1991, pp. 368--377.

....71 Learning with Missing Attribute Values: Every learning algorithm should handle missing attribute values in some way, because most of the real world datasets contain unknown attribute values. Therefore, in the literature, there are some methods to handle these kinds of attribute values [23, 43, 44, 46]. Most of the methods are based on one of the following ideas: ffl Ignoring examples that have unknown attribute value. ffl Assuming an additional special value for unknown attribute values. ffl Using probability theory by utilizing information provided by context. ffl Generating additional ....

....dimensions are independent from each other, no specialization is required. The concept descriptions can be overlapped. Another important property of the COFI algorithm is its way of handling the unknown attribute values. Most of the systems use ad hoc methods to handle the unknown attribute values [23, 44]. Like CFP, the COFI algorithm also ignores the unknown attribute values. Since the value of each attribute is handled separately, this causes no problem. The behavior of the COFI algorithm to the irrelevant features is very interesting. Irrelevant attributes can easily be detected by looking at ....

J.W. Grzymala-Busse, On the Unknown Attribute Values in Learning from Examples, In Proceedings of Sixth International Symposium Methodologies for Intelligent Systems, pp: 368-377, October 1991.

....inapplicable for some tuples. For example, in the list of personal computers, the attribute that contains the model type of the sound cards would be null for some model of computers. We have not come across any work that deals with null values, though there are some recent studies on unknown values[24, 11, 34]. 4. Incomplete or Redundant Data: The fact that data has been organized and collected around the needs of organizational activities causes incomplete or redundant data from the view point of the knowledge discovery task. The situation of incomplete data arises when the available information based ....

....value of that attribute, and the reasoning about null values remains an open problem in the studies of database mining. A less restrictive version of the problem, which is known as unknown attribute values, has been studied by Grzymala Busse and implemented in the LERS, a machine learning system[11, 12]. ffl Characterization query: Even though data dependency analysis within the rough set methodology can be applied to characterize concepts, it lacks an explicit structure such as hierarchy of persistent concepts to exploit concept dependencies. This subject has been formally studied by ....

J. W. Grzymala-Busse. On the unknown attribute values in learning from examples. In Z. W. Ras and M. Zemankowa, editors, Proceedings of Methodologies for Intelligent Systems, Lecture Notes in AI, 542, pages 368--377. Springer-Verlag, New York, 1991.

....classification noise, attribute noise etc. that can be found in real world data sets. Many researchers tackled this problem [3, 12, 35] Another type of noise is the unknown (missing) attribute values. In order to cope with missing attribute values, many techniques were tried. For example, in [15] additional instances are generated for all possible values of the missing attribute and rough set theory is used to solve the conflicts. Obviously, this approach is only applicable to attributes that have finite number of possible values. However, it is a costly solution to handle unknown ....

....threshold cause removing partitions more aggressively. If confidence threshold is zero then percentage of the noise in the concept description is equal to the noise level of the training set. Unknown Attribute Values Most of the real world data sets contain missing attribute values. Many authors [15, 27, 28, 29] were presented methods for handling unknown attribute values. Most of the methods are based on the following ideas: 1. Ignoring examples which have unknown attribute value. 2. Assuming an additional special value for unknown attribute values. This can lead to an anomalous situation [27] 3. ....

[Article contains additional citation context not shown here]

J. W. Grzymala-Busse. On the Unknown Attribute Values in Learning from Examples. In Proceedings of Sixth International Symposium Methodologies for Intelligent Systems, pages 368--377, October 1991.

....is handled by subdividing it into three cases such as unknown, inapplicable, and unknown or inapplicable. Other than this work, which does not offer any solution for existing data, we have not come across any work that deals with null values, though there are some recent studies on unknown values [28, 29, 30]. When the database contains missing attribute values, either the values can be discarded or an attempt can be made to replace them with the most likely values [19] This were the ideas adopted by Quinlan [19] for inductive decision trees. In [31] it was suggested to construct rules that predict ....

....rules that predict the value of the missing attribute, based on the value of other attributes in the example, and the class information. These rules can then be used to fill in the missing attribute values and the resulting data set could be used to construct the descriptions. Grzymala Busse [29], citing the drawbacks of approaches given above, has transformed a given decision table with unknown values to a new and possibly inconsistent decision table, in which every attribute value is known, by replacing the unknown value of an attribute with all possible values of that attribute. In ....

J. W. Grzymala-Busse, "On the unknown attribute values in learning from examples," in Proceedings of Methodologies for Intelligent Systems (Z. W. Ras and M. Zemankowa, eds.), Lecture Notes in AI, 542, pp. 368--377, New York: Springer-Verlag, 1991.

....noise. 5.2.3.3 Experiments with Increasing Level of Missing Values Most of the real world datasets contain missing (unknown) feature values and the percentage of missing values are shown in Table A.1. In order to cope with instances that contain missing values, several methods have been proposed [30, 55, 56, 57, 58]. These methods can be summarized as: ffl Ignoring instances which have unknown feature values. ffl Assuming an additional special value for unknown attribute values. ffl Using probability theory by utilizing information provided by context. ffl Generating additional instances for all possible ....

.... OF THE LEARNED CONCEPTS 134 Feature values of test instance 1: F[1] 2 F[2] 3 F[3] 3 F[4] 3 F[5] 3 F[6] 0 F[7] 0 F[8] 0 F[9] 3 F[10] 3 F[11] 0 F[12] 0 F[13] 0 F[14] 0 F[15] 0 F[16] 0 F[17] 3 F[18] 2 F[19] 2 F[20] 3 F[21] 3 F[22] 3 F[23] 1 F[24] 3 F[25] 0 F[26] 0 F[27] 0 F[28] 0 F[29] 0 F[30]:0 F[31] 0 F[32] 1 F[33] 0 F[34] 34 Classes: 1] 2] 3] 4] 5] 6] Votes of Feature[1] 0.16 0.16 0.17 0.18 0.15 0.18 Votes of Feature[2] 0.36 0.27 0.15 0.07( 0.05( 0.10 Votes of Feature[3] 0.34 0.05( 0.38 0.07( ....

[Article contains additional citation context not shown here]

J.W. Grzymala--Busse, On the Unknown Attribute Values in Learning from Examples, In Proceedings of Sixth International Symposium Methodologies for Intelligent Systems, 368--377, October 1991. BIBLIOGRAPHY 150

....is handled by subdividing it into three cases such as unknown, inapplicable, and unknown or inapplicable. Other than this work, which does not offer any solution for existing data, we have not come across any work that deals with null values, though there are some recent studies on unknown values [28, 29, 30]. When the database contains missing attribute values, either the values can be discarded or an attempt can be made to replace them with the most likely values [19] These are the ideas adopted by Quinlan [19] for inductive decision trees. In [31] it is suggested to construct rules that predict ....

....the value of the missing attribute, Data Mining 9 based on the value of other attributes in the example, and the class information. These rules can then be used to fill in the missing attribute values and the resulting data set could be used to construct the descriptions. Grzymala Busse [29], citing the drawbacks of the approaches given above, has transformed a given decision table with unknown values to a new and possibly inconsistent decision table, in which every attribute value is known, by replacing the unknown value of an attribute with all possible values of that attribute. In ....

J. W. Grzymala-Busse, "On the unknown attribute values in learning from examples, " in Proceedings of Methodologies for Intelligent Systems (Z. W. Ras and M. Zemankowa, eds.), Lecture Notes in AI, 542, pp. 368--377, New York: SpringerVerlag, 1991. 34 Chapter 1

....it may not be computationally feasible to find a reduct. Furthermore, finding just a single reduct may be too restrictive for some data analysis problems. One plausible approach is to utilize the idea of reduct defined in the previous subsection[6] To handle missing values, Grzymala Busse [8] has transformed a given decision table with unknown values to a new and possibly inconsistent decision table by replacing the unknown attribute value with all possible values of that attribute. In other words, he reduced the missing value problem to that of learning from inconsistent examples. He ....

GRZYMALA-BUSSE, J. W. On the unknown attribute values in learning from examples. In Proceedings of Methodologies for Intelligent Systems, Z. W. Ras and M. Zemankowa, Eds., Lecture Notes in AI, 542. Springer-Verlag, New York, 1991, pp. 368--377.

No context found.

Grzymala-Busse, J.W. 1991. On the Unknown Attribute Values in Learning from Examples.

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