A General Framework for Fuzzy Data Mining

Pemerintah Indonesia menggunakan Nilai Ujian Nasional (NUN) digunakan sebagai standar nasional untuk mengukur keberhasilan siswa. Dengan menggunakan NUN sebagai parameter untuk mengukur tingkat keberhasilan siswa akan ada resiko yang sangat besar jika siswa gagal dalam menjalani Ujian Nasional (UN), oleh karena itu dibutuhkan sebuah model untuk melakukan prediksi terhadap NUN siswa sehingga dapat dilakukan usaha pencegahan terhadap gagalnya siswa dalam UN. Pada paper ini diusulkan sebuah metode untuk melakukan prediksi terhadap NUN siswa dengan menggunakan neuro fuzzy. Data yang digunakan dalam penelitian ini adalah nilai rapor siswa, dan nilai IQ siswa. Pada penelitian ini digunakan tiga mata pelajaran sebagai subjek penelitian yaitu Matematika, Bahasa Indonesia dan Bahasa Inggris.Berdasarkan hasil pengujian yang dilakukan menunjukkan bahwa metode neuro fuzzy dengan menggunakan data nilai dan data IQmemberikan akurasi prediksi terbaik pada mata pelajaran matematika sebesar 74%.

Download Full-text

Finding Active Membership Functions for Genetic-Fuzzy Data Mining

International Journal of Information Technology & Decision Making ◽

10.1142/s0219622015500297 ◽

2015 ◽

Vol 14 (06) ◽

pp. 1215-1242 ◽

Cited By ~ 3

Author(s):

Chun-Hao Chen ◽

Tzung-Pei Hong ◽

Yeong-Chyi Lee ◽

Vincent S. Tseng

Keyword(s):

Data Mining ◽

Association Rules ◽

Fuzzy Data ◽

Membership Functions ◽

Linguistic Terms ◽

Fuzzy Association Rules ◽

Fitness Functions ◽

The Given ◽

Binary Strings ◽

Active Membership

Since transactions may contain quantitative values, many approaches have been proposed to derive membership functions for mining fuzzy association rules using genetic algorithms (GAs), a process known as genetic-fuzzy data mining. However, existing approaches assume that the number of linguistic terms is predefined. Thus, this study proposes a genetic-fuzzy mining approach for extracting an appropriate number of linguistic terms and their membership functions used in fuzzy data mining for the given items. The proposed algorithm adjusts membership functions using GAs and then uses them to fuzzify the quantitative transactions. Each individual in the population represents a possible set of membership functions for the items and is divided into two parts, control genes (CGs) and parametric genes (PGs). CGs are encoded into binary strings and used to determine whether membership functions are active. Each set of membership functions for an item is encoded as PGs with real-number schema. In addition, seven fitness functions are proposed, each of which is used to evaluate the goodness of the obtained membership functions and used as the evolutionary criteria in GA. After the GA process terminates, a better set of association rules with a suitable set of membership functions is obtained. Experiments are made to show the effectiveness of the proposed approach.

Download Full-text

Neuro-fuzzy Data Mining for Target Group Selection in Retail Banking

Principles of Data Mining and Knowledge Discovery - Lecture Notes in Computer Science ◽

10.1007/978-3-540-48247-5_69 ◽

1999 ◽

pp. 530-535

Author(s):

Johannes Ruhland ◽

Thomas Wittmann

Keyword(s):

Data Mining ◽

Group Selection ◽

Retail Banking ◽

Target Group ◽

Fuzzy Data ◽

Neuro Fuzzy

Download Full-text

Handling Fuzzy Similarity for Data Classification

Encyclopedia of Artificial Intelligence ◽

10.4018/978-1-59904-849-9.ch118 ◽

2011 ◽

pp. 796-802 ◽

Cited By ~ 2

Author(s):

Roy Gelbard ◽

Avichai Meged

Keyword(s):

Data Mining ◽

Similarity Measure ◽

Data Representation ◽

Classification Algorithms ◽

Binary Representation ◽

Fuzzy Data ◽

Problem Domain ◽

Data Types ◽

Fuzzy Similarity ◽

Set Up

Representing and consequently processing fuzzy data in standard and binary databases is problematic. The problem is further amplified in binary databases where continuous data is represented by means of discrete ‘1’ and ‘0’ bits. As regards classification, the problem becomes even more acute. In these cases, we may want to group objects based on some fuzzy attributes, but unfortunately, an appropriate fuzzy similarity measure is not always easy to find. The current paper proposes a novel model and measure for representing fuzzy data, which lends itself to both classification and data mining. Classification algorithms and data mining attempt to set up hypotheses regarding the assigning of different objects to groups and classes on the basis of the similarity/distance between them (Estivill-Castro & Yang, 2004) (Lim, Loh & Shih, 2000) (Zhang & Srihari, 2004). Classification algorithms and data mining are widely used in numerous fields including: social sciences, where observations and questionnaires are used in learning mechanisms of social behavior; marketing, for segmentation and customer profiling; finance, for fraud detection; computer science, for image processing and expert systems applications; medicine, for diagnostics; and many other fields. Classification algorithms and data mining methodologies are based on a procedure that calculates a similarity matrix based on similarity index between objects and on a grouping technique. Researches proved that a similarity measure based upon binary data representation yields better results than regular similarity indexes (Erlich, Gelbard & Spiegler, 2002) (Gelbard, Goldman & Spiegler, 2007). However, binary representation is currently limited to nominal discrete attributes suitable for attributes such as: gender, marital status, etc., (Zhang & Srihari, 2003). This makes the binary approach for data representation unattractive for widespread data types. The current research describes a novel approach to binary representation, referred to as Fuzzy Binary Representation. This new approach is suitable for all data types - nominal, ordinal and as continuous. We propose that there is meaning not only to the actual explicit attribute value, but also to its implicit similarity to other possible attribute values. These similarities can either be determined by a problem domain expert or automatically by analyzing fuzzy functions that represent the problem domain. The added new fuzzy similarity yields improved classification and data mining results. More generally, Fuzzy Binary Representation and related similarity measures exemplify that a refined and carefully designed handling of data, including eliciting of domain expertise regarding similarity, may add both value and knowledge to existing databases.

Download Full-text