Maimon, O., Kandel, A. & Last, M. (2001). Information-Theoretic Fuzzy Approach to Data Reliability and Data Mining. Fuzzy Sets and Systems. Vol. 117, No. 2, pp. 183194.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....numeric values, which do not possess any meaning in the original categorical attributes, are used to induce clusters. The associations concerned with these attributes discovered by this method may therefore be misleading. Furthermore, an information theoretic fuzzy approach has been proposed in [13] to discover unreliable data in databases. This approach first defines some attributes as input or target attributes in a given relational database and then constructs a connectionist network to evaluate the reliability of values of target attributes in every record as a fuzzy measure. Unreliable ....

O. Maimon, A. Kandel, and M. Last, "InformationTheoretic Fuzzy Approach to Data Reliability and Data Mining," Fuzzy Sets and Systems, vol. 117, pp. 183-194, 2001.

....[27] express knowledge in linguistic representation, which is natural for people to comprehend. In addition to linguistic summaries, the applicability of fuzzy modeling techniques to data mining has been discussed in [13] Furthermore, an information theoretic fuzzy approach has been proposed in [17] to discover unreliable data in databases. Nevertheless, these fuzzy techniques have not been developed for classification. An approach, which combines symbolic decision trees with approximate reasoning offered by fuzzy representation, has been proposed in [14] for building fuzzy decision trees. ....

O. Maimon, A. Kandel, and M. Last, "InformationTheoretic Fuzzy Approach to Data Reliability and Data Mining," Fuzzy Sets and Systems, vol. 117, pp. 183-194, 2001.

.... Automated Detection of Outliers in Real World Data Mark Last Department of Information Systems Engineering Ben Gurion University of the Negev Beer Sheva 84105, Israel E mail: mlast bgumail.bgu.ac.il Abraham Kandel Department of Computer Science and Engineering University of South Florida 4202 E. Fowler Avenue, ENB 118 Tampa, FL 33620, USA E mail: kandel csee.usf.edu Abstract: Most real world databases include a certain amount of exceptional ....

....of Outliers in Real World Data Mark Last Department of Information Systems Engineering Ben Gurion University of the Negev Beer Sheva 84105, Israel E mail: mlast bgumail.bgu.ac.il Abraham Kandel Department of Computer Science and Engineering University of South Florida 4202 E. Fowler Avenue, ENB 118 Tampa, FL 33620, USA E mail: kandel csee.usf.edu Abstract: Most real world databases include a certain amount of exceptional values, generally termed as outliers . The isolation of outliers is important both for improving the quality of original data and for reducing the impact of outlying ....

[Article contains additional citation context not shown here]

Maimon, O., Kandel, A. & Last, M. (2001). Information-Theoretic Fuzzy Approach to Data Reliability and Data Mining. Fuzzy Sets and Systems. Vol. 117, No. 2, pp. 183194.

....estimate quickly, and with a high degree of confidence, the reliability of obtained information. He, or she, would consider it as highly reliable , not so reliable , doubtful , absolutely unreliable , etc. To automate the human perceptions of reliable and unreliable data, we have suggested in [10] the following definition of data reliability: Degree of Reliability of an attribute A in a record k is defined on a unit interval [0,1] as the degree of certainty that the value of attribute A stored in a record k is correct from user s point of view. After applying a data mining algorithm to a ....

....10 20 30 40 50 60 70 80 90 100 00.511.522.533.544.55 Distance (d) Reliability Degree 1 5 Fig. 1 Reliability perceptions for different values of beta d ik a measure of distance between the predicted value j and the actual value j of the attribute A in the record k. In [10], we use the following distance function: log z V P z V P d ij ij ik = 2) where ) z V P ij estimated probability of the predicted value j , given a model z; z V P ij estimated probability of the actual value j, given a model z. The expression ....

[Article contains additional citation context not shown here]

O. Maimon, A. Kandel, and M. Last, InformationTheoretic Fuzzy Approach to Data Reliability and Data Mining, to appear in Fuzzy Sets and Systems, 1999.

....sentence are independent, then the two layer network reduces to the single layer network, but they do not propose any analytical way of selecting the number of layers (analogous to the number of attributes in a database) in a practical case. The learning method of [7 8] has been enhanced by us in [13 14] for the task of evaluating reliability of partially reliable attributes in a database as a fuzzy logic measure. In [13 14] we have used a constant number of network layers. This work extends the method of [13 14] to the problem of finding the minimal number of network layers representing a ....

....way of selecting the number of layers (analogous to the number of attributes in a database) in a practical case. The learning method of [7 8] has been enhanced by us in [13 14] for the task of evaluating reliability of partially reliable attributes in a database as a fuzzy logic measure. In [13 14] we have used a constant number of network layers. This work extends the method of [13 14] to the problem of finding the minimal number of network layers representing a minimum set of significant input attributes. B. Paper Contribution The contributions of our approach include the following: ....

[Article contains additional citation context not shown here]

O. Maimon, A. Kandel, M. Last, Information-Theoretic Fuzzy Approach to Data Reliability and Data Mining, Submitted to Publication, 1997.

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