Search

We consider fuzzy inference of the form ‘if ··· then ··· ’. ‘if ··· then ··· else ···’ and ‘if ··· and/or ··· and/or ··· then ···’. Muzomoto proposed logical constructs for ‘if. then.’ by applying ...

Data mining is knowledge discovery process. It has to deal with exact information and inexact information. Statistical methods deal with inexact information but it is based on likelihood. Zadeh fuzzy logic deals with inexact information but it is based on belief and it is simple to use. Fuzzy logic is used to deal with inexact information. Data mining consist methods and classifications. These methods and classifications are discussed for both exact and inexact information. Retrieval of information is important in data mining. The time and space complexity is high in big data. These are to be reduced. The time complexity is reduced through the consecutive retrieval (C-R) property and space complexity is reduced with blackboard systems. Data mining for web data based is discussed. In web data mining, the original data have to be disclosed. Fuzzy web data mining is discussed for security of data. Fuzzy web programming is discussed. Data mining, fuzzy data mining, and web data mining are discussed through MapReduce algorithms.

In a distributed environment relations are stored at different sites. To perform algebraic operations such as join, the relations are to be transferred from one site to the other in such a way that the total communication cost is minimized. This paper deals with the problem of computing the transmission cost using two approaches. The first uses System R* algorithm approach when the data is of non-skew nature and the second uses SharesSkew algorithm when the data has skews i.e., same value for a specific join attribute, named as Heavy Hitter (HH). Rules of the two algorithms to be followed for performing join are specified and by illustrating with Banking System, the communication cost is evaluated.

Defining data inherently with fuzziness reduce the complexity of data mining during knowledge discovery process. The mining with fuzzy database will provide security because the original data need not be disclosed The fuzzy logic with two membership functions will give more evidence than the Zadeh single membership function. In this paper, the fuzzy data mining methods are discussed. The two fold fuzzy set with two membership functions is studied with "Belief" and ""Disbelief". The fuzzy certainty factor (FCF) is difference of the two membership functions to eliminate conflict, The FCF and gives single fuzzy membership function. The fuzzy risk set is defined with fuzzy certainty factor for decision making. The fuzzy MapReducing with functional dependency is studied for association rules. The Generalized fuzzy reasoning is studied for data mining. The data mining with fuzzy risk set is studied to take the decisions. The business intelligence is given as an example.

The amount of data being stored in various repositories has been growing rapidly. In order to uncover hidden information from such data, there is a need for information extraction techniques. This resulted in the use of Data Mining and Machine Learning tasks such as Classification. One among the industries producing huge number of records on a daily basis is the Healthcare system. Data associated with medical field deal with the diagnosis of patients, higher accuracy in decision making is expected. In this work, we deal with the analysis of Cleveland Heart Disease data which contains 303 instances of patients records. Four classifiers are applied on the processed data to measure the accuracy of classification. Our results show that Deep Neural Net MLP classifier is able to diagnose the status of a patient with more accuracy.

Data mining is knowledge discovery process. It has to deal with exact information and inexact information. Statistical methods deal with inexact information but it is based on likelihood. Zadeh fuzzy logic deals with inexact information but it is based on belief and it is simple to use. Fuzzy logic is used to deal with inexact information. Data mining consist methods and classifications. These methods and classifications are discussed for both exact and inexact information. Retrieval of information is important in data mining. The time and space complexity is high in big data. These are to be reduced. The time complexity is reduced through the consecutive retrieval (C-R) property and space complexity is reduced with blackboard systems. Data mining for web data based is discussed. In web data mining, the original data have to be disclosed. Fuzzy web data mining is discussed for security of data. Fuzzy web programming is discussed. Data mining, fuzzy data mining, and web data mining are discussed through MapReduce algorithms.

Zadeh proposed fuzzy logic with single membership function. Two Zadeh, Mamdani and TSK proposed fuzzy conditional inference. In many applications like fuzzy control systems, the consequent part may be derived from precedent part. Zadeh, Mamdani and TSK proposed different fuzzy conditional inferences for “if … then …” for approximate reasoning. The Zadeh and Mamdani fuzzy conditional inferences are know prior information for both precedent part and consequent part. The TSK fuzzy conditional inferences need not know prior information for consequent part but it is difficult to compute. In this chapter, fuzzy conditional inference is proposed for “if…then…” This fuzzy conditional inference need not know prior information of the consequent part. The fuzzy conditional inference is discussed using the single fuzzy membership function and twofold fuzzy membership functions. The fuzzy control system is given as an application.

Zadeh, Mamdani, TSK and Fukami are proposed different fuzzy conditional inferences. Mamdani is also proposed for nested fuzzy conditional inference. In this paper, some methods on fuzzy conditional inference and fuzzy reasoning are studied to approximate reasoning. The Business Intelligence and Medical diagnosis are given as an example.

the logical data independence and physical data independence are necessary for data mining particularly for BIG DATA. The data base defining inherency with fuzziness will reduce the time in Information retrieval. The Data Mining is very fast using the fuzzy log for BIG DATA. In this paper, fuzzy data mining is studied and fuzzy association dependency is defied. The fuzzy MapReducibg algorithm is studied for BIG DATA to reduce the retrieval time for BIG DATA. This algorithm wills qick reference to perticular information and actual information will be secured.

Knowledge representation is key factor for problem representation particularly for incomplete information in expert systems. Usually incomplete information is fuzzy rather than likelihood. Learning methods are necessary to solve expert problems. The conditional inference method is studied different from Zadeh, Mamdani and TSK methods. Learning fuzzy conditional inference with individual methods fuzzy logic, Genetic algorithms, Petri net and neural net not sufficient for large problems. In this paper FuGePeNuNet method is studded by combining fuzzy logic, Genetic algorithms, Petri net and Neural net for large problems of Fuzzy Expert Systems. The fuzzy medical diagnose is given an example.