FILM: a fuzzy inductive learning method for automated knowledge acquisition

doi:10.1016/S0167-9236(97)00019-5

Decision Support Systems

Volume 21, Issue 2, October 1997, Pages 61-73

https://doi.org/10.1016/S0167-9236(97)00019-5 Get rights and content

Abstract

Inductive learning that creates decision trees from a set of existing cases is useful for automated knowledge acquisition. Most of the existing methods in literature are based on crisp concepts that are weak in handling marginal cases. In this paper, we present a fuzzy inductive learning method that integrates the fuzzy set theory into the regular inductive learning processes. The method converts a decision tree induced from regular method into a fuzzy decision tree in which hurdle values for splitting branches and classes associated with leaves are fuzzy. Results from empirical tests indicate that the new fuzzy approach outperforms the popular discriminant analysis and ID3 in predictive accuracy.

References (24)

J.R. Quinlan
Simplifying decision trees
Int. J. Man-Machine Studies
(1987)
L.A. Zadeh
Fuzzy sets
Inf. Control
(1965)
B. Jeng et al.
Fuzzy indexing and retrieval in case-based systems
Expert Systems with Applications
(1995)
T.P. Liang et al.
Integrating statistical and inductive learning methods for knowledge acquisition
Expert Systems with Applications
(1990)
H. Braun et al.
Predicting stock market behavior through rule induction: an application of the learning-from-example approach
Decision Sci.
(1987)
C. Carter et al.
Assessing Credit Card Applications Using Machine Learning
IEEE Expert
(1987)
H.M. Chung et al.
Rule-based expert systems and linear models: an empirical comparison of learning-by-examples methods
Decision Sci.
(1992)
T.P. Liang et al.
An Empirical investigation of some data effects on the classification accuracy of probit, ID3 and neural networks
Contemp. Accounting Res.
(1992)
W.F. Messier et al.
Inducing rules for expert system development: an example using default and bankruptcy data
Manage. Sci.
(1988)
M.J. Shaw et al.
Using an expert system with inductive learning to evaluate business loans
Financial Manage.
(1988)

A. Marchand et al.

The assessment of laboratory tests in the diagnosis of Acute appendicitis

Am. J. Clin. Pathol.

(1983)

R. Michalski et al.

The Multi-purpose Incremental Learning System AQ15 and Its Testing Application to Three Medical Domains

Cited by (47)

Systematic review of bankruptcy prediction models: Towards a framework for tool selection
2018, Expert Systems with Applications
Citation Excerpt :
DT induces the decision rules. The positions of the rules in the decision tree are usually determined using heuristics (Jeng, Jeng, & Liang, 1997). For example, if profitability was found to be more important than liquidity, it will be placed above, or evaluated before, liquidity.
The bankruptcy prediction research domain continues to evolve with many new different predictive models developed using various tools. Yet many of the tools are used with the wrong data conditions or for the wrong situation. Using the Web of Science, Business Source Complete and Engineering Village databases, a systematic review of 49 journal articles published between 2010 and 2015 was carried out. This review shows how eight popular and promising tools perform based on 13 key criteria within the bankruptcy prediction models research area. These tools include two statistical tools: multiple discriminant analysis and Logistic regression; and six artificial intelligence tools: artificial neural network, support vector machines, rough sets, case based reasoning, decision tree and genetic algorithm. The 13 criteria identified include accuracy, result transparency, fully deterministic output, data size capability, data dispersion, variable selection method required, variable types applicable, and more. Overall, it was found that no single tool is predominantly better than other tools in relation to the 13 identified criteria. A tabular and a diagrammatic framework are provided as guidelines for the selection of tools that best fit different situations. It is concluded that an overall better performance model can only be found by informed integration of tools to form a hybrid model. This paper contributes towards a thorough understanding of the features of the tools used to develop bankruptcy prediction models and their related shortcomings.
Review on various modelling techniques for the solar dryers
2016, Renewable and Sustainable Energy Reviews
This review paper is focused on the various modelling techniques for the solar dryer system. The modelling techniques are very important to develop, increase drying efficiency, analyse and predict the performance of different kinds of solar drying system. The modelling techniques are also important for predicting the temperature of crop moisture content, drying rate, quality of crop and colour of crops. Computational fluid dynamics (CFD can be applied for analysing and investigating of air flow and spry of temperature in the drying system. Adaptive-network-based fuzzy inference system (ANFIS) can be used to predict the behaviour of the solar drying system. ANN is used to calculate the mass of the dried crops on hourly basis. FUZZY is very important software for using the simulation of drying system. That can also be used to accurately predict the results with a minimum error. The mathematical modelling techniques are used for testing the drying behaviour of crops in the laboratory. It act in effect tool between scientists and investigators. It helps short of spending vast amount of time, energy and money in experimental events. Before fabrication the modelling techniques are very supportive in simulation of different types of solar drying system. Thus, analysis on the base of modelling techniques is not only save time but also save the capital investment in solar drying system. The advantage and future scope of modelling techniques is also discussed.
Predicting restaurant financial distress using decision tree and AdaBoosted decision tree models
2014, Economic Modelling
Citation Excerpt :
Specially, the non-parametric prediction method known as decision tree (DT) or recursive partitioning has been used in an attempt to bypass the above mentioned assumptions in MDA and logit (Frydman et al., 1985; Marais et al., 1984). In addition to the previous DT studies (e.g., Jeng et al., 1997; Lee et al., 1996; Olmeda and Fernandez, 1997; Tam and Kiang, 1992), more recent studies used DT in financial distress prediction (Bastos, 2008; Gepp et al., 2010; Huarng et al., 2005; Joos et al., 1998; Koh, 2004; Li et al., 2010; Lin and McClean, 2001; Quinlan, 1996; Shirata, 1998). Past research has justified the search for the new financial distress prediction approaches of DT models and Adaptive boosted (AdaBoosted) DT models in the context of the restaurant industry.
The restaurant industry has been facing tough challenges because of the recent economic turmoil. Although different industries face different levels of competition and therefore the likelihood of financial distress can differ for firms in different industries, scant attention has been paid to predicting restaurant financial distress. The primary objective of this paper is to examine the key financial distress factors for publicly traded U.S. restaurants for the period from 1988 to 2010 using decision trees (DT) and AdaBoosted decision trees. The AdaBoosted DT model for the entire dataset revealed that financially distressed restaurants relied more heavily on debt; and showed lower rates of increase of assets, lower net profit margins, and lower current ratios than non-distressed restaurants. A larger proportion of debt in the capital structure ruined restaurants' financial structure and the inability to pay their drastically increased debt exposed restaurants to financial distress. Additionally, a lack of capital efficiency increased the possibility of financial distress. We recommend the use of the AdaBoosted DT model as an early warning system for restaurant distress prediction because the AdaBoosted DT model demonstrated the best prediction performance with the smallest error in overall and type I error rates. The results of two subset models for full-service and limited-service restaurants indicated that the segments had slightly different financial risk factors.
Decision tree based fuzzy classifier of <sup>1</sup>H magnetic resonance spectra from cerebrospinal fluid samples
2011, Fuzzy Sets and Systems
This paper presents a method for classifying cerebrospinal fluid (CSF) samples studied by proton magnetic resonance spectroscopy ( ${}^{1}H$ MRS) into clinical subgroups by means of a fuzzy classifier. The method focuses on the analysis of a low signal-to-noise region of the spectra and is designed to use a small number of samples because sampling can only be done through an invasive technique. The proposed method involves the fusion of classifiers based on decision trees designed using fuzzy techniques. The fusion step was carried out by ordered weighted averaging (OWA) operators. The quality of the proposed classifier was evaluated by efficiency and robustness quality indexes using a method based on a cross-validation technique. Results show excellent classification levels and satisfactory robustness in both training and test sets.
Recognition of multi-interval rules in dataset with continuous-valued attributes
2009, Expert Systems with Applications
Citation Excerpt :
By such induced knowledge structure, the system can therefore deal with similar tasks with more efficiency and effectiveness (Simon, 1983). Knowledge retrieval based on machine learning is a process which turns human knowledge into machine code that can be recognized by computers (Jeng, Jeng, & Liang, 1997). The knowledge structures learned by machine learning are more convenient to manage through ordinal methods (Pazzani & Kibler, 1992).
The decision tree induction learning approach in machine learning has been extensively applied to the field of knowledge management in practice, since it provides the advantages of quick learning and easy creation of explicit knowledge structures. However, the congenital limitation of node and branch structure may limit the success of decision tree induction learning in dealing with nominal or discrete-valued attributes. The need for discretizing or splitting a candidate continuous-valued attribute into some finite manageable number of intervals therefore is crucial in the application of decision tree induction learning. In this study, an integrated approach was proposed by utilizing the decision tree induction along with the hierarchical clustering analysis which combined the appropriate intervals based on the use of the proposed measure with considerations of both within attribute closeness and discretization results similarity. This proposed integrated approach facilitates the task of multi-interval discretization to produce more accurate classification rules by improving the processing difficulty of continuous-valued attributes for decision tree induction learning. Finally, the proposed approach was tested in terms of both the predictive accuracy and the size of the decision tree by using the UCI-ML testing databases.
A methodology for automated fuzzy model generation
2008, Fuzzy Sets and Systems
In this paper we propose a generic methodology for the automated generation of fuzzy models. The methodology is realized in three stages. Initially, a crisp model is created and in the second stage it is transformed to a fuzzy one. In the third stage, all parameters entering the fuzzy model are optimized. The proposed methodology is novel and generic since it can integrate alternative techniques in each of its stages. A specific realization of this methodology is implemented, using decision trees for the creation of the crisp model, the sigmoid function, the min–max operators and the maximum defuzzifier, for the transformation of the crisp model into a fuzzy one, and four different optimization strategies, including global and local optimization techniques, as well as, hybrid approaches. The proposed methodology presents several advantages and novelties: the transformation of the crisp model to the respective fuzzy one is straightforward ensuring its full automated nature and it introduces a set of parameters, expressing the importance of each fuzzy rule. The above realization is extensively evaluated using several benchmark data sets from the UCI machine learning repository and the obtained results indicate its high efficiency.

View all citing articles on Scopus

View full text

PaperFILM: a fuzzy inductive learning method for automated knowledge acquisition

Abstract

Int. J. Man-Machine Studies

Inf. Control

Expert Systems with Applications

Expert Systems with Applications

Predicting stock market behavior through rule induction: an application of the learning-from-example approach

Decision Sci.

Assessing Credit Card Applications Using Machine Learning

IEEE Expert

Rule-based expert systems and linear models: an empirical comparison of learning-by-examples methods

Decision Sci.

An Empirical investigation of some data effects on the classification accuracy of probit, ID3 and neural networks

Contemp. Accounting Res.

Inducing rules for expert system development: an example using default and bankruptcy data

Manage. Sci.

Using an expert system with inductive learning to evaluate business loans

Financial Manage.

The assessment of laboratory tests in the diagnosis of Acute appendicitis

Am. J. Clin. Pathol.

The Multi-purpose Incremental Learning System AQ15 and Its Testing Application to Three Medical Domains

Paper
FILM: a fuzzy inductive learning method for automated knowledge acquisition