Alex Tormási

55389496800

Publications - 15

Two Stages Outlier Removal as Pre-Processing Digitizer Data on Fine Motor Skills (FMS) Classification Using Covariance Estimator and Isolation Forest

Surya Sumpeno Alex Tormási Nurul Zainal Fanani Festa Yumpi Rahmanawati K. Khamid Adri Gabriel Sooai Mauridhi Hery Purnomo L. T. Koczy

Publication Name: International Journal of Intelligent Engineering and Systems

Publication Date: 2021-08-01

Volume: 14

Issue: 4

Page Range: 571-582

Description:

The increase of the classification accuracy level has become an important problem in machine learning especially in diverse data-set that contain the outlier data. In the data stream or the data from sensor readings that produce large data, it allows a lot of noise to occur. It makes the performance of the machine learning model is disrupted or even decreased. Therefore, clean data from noise is needed to obtain good accuracy and to improve the performance of the machine learning model. This research proposes a two-stages for detecting and removing outlier data by using the covariance estimator and isolation forest methods as pre-processing in the classification process to determine fine motor skill (FMS). The dataset was generated from the process of recording data directly during cursive writing by using a digitizer. The data included the relative position of the stylus on the digitizer board. x position, y position, z position, and pressure values are then used as features in the classification process. In the process of observation and recording, the generated data was very huge so some of them produce the outlier data. From the experimental results that have been implemented, the level of accuracy in the FMS classification process increases between 0.5-1% by using the Random Forest classifier after the detection and outlier removal by using covariance estimator and isolation forest. The highest accuracy rate achieves 98.05% compared to the accuracy without outlier removal, which is only about 97.3%.

Open Access: Yes

DOI: 10.22266/ijies2021.0831.50

A rule-based expert system for automatic question classification in mathematics adaptive assessment on indonesian elementary school environment

Siti Rochimah Umi Laili Yuhana Alex Tormási Eko Mulyanto Yuniarno Aliia Rysbekova Mauridhi Hery Purnomo L. T. Koczy

Publication Name: International Journal of Innovative Computing Information and Control

Publication Date: 2019-02-01

Volume: 15

Issue: 1

Page Range: 143-161

Description:

This paper is part of research in developing a competency-based assessment system for mathematics in Indonesian elementary school environment. An essential task is to accurately classify questions based on competency and difficulty level. Thus, an expert system is needed to classify those questions since competency information is often manually defined by experts. The objectives of this work are replacing a human expert’s role in the related knowledge engineering process and providing a rule-based expert system to supersede an expert to classify the questions. Five types of the rule-based algorithm: OneR, RIPPER, PART, FURIA, and J48, were applied to the dataset, which is comprised of 9454 real mathematics examination questions collected from several Indonesian elementary schools. Following the knowledge engineering principles, these algorithms generated the classification rules based on a pattern of the data. The rules of the best performing algorithm were utilized by a knowledge base for inference. Finally, to be able to fully measure the system performance, ten expert teachers were involved in the question classification step. The results confirm that the system meets the stated objectives in classifying the competency and the difficulty level of a question automatically.

Open Access: Yes

DOI: 10.24507/ijicic.15.01.143

Experimenting with a new population-based optimization technique: FUNgal growth inspired (FUNGI) optimizer

Alex Tormási L. T. Koczy

Publication Name: Studies in Fuzziness and Soft Computing

Publication Date: 2018-01-01

Volume: 361

Issue: Unknown

Page Range: 123-135

Description:

In this paper the experimental results of a new evolutionary algorithm are presented. The proposed method was inspired by the growth and reproduction of fungi. Experiments were executed and evaluated on discretized versions of common functions, which are used in benchmark tests of optimization techniques. The results were compared with other optimization algorithms and the directions of future research with many possible modifications/extension of the presented method are discussed.

Open Access: Yes

DOI: 10.1007/978-3-319-75408-6_11

A survey of the applications of fuzzy methods in recommender systems

P. Földesi Alex Tormási Brigita Sziová L. T. Koczy

Publication Name: Studies in Fuzziness and Soft Computing

Publication Date: 2018-01-01

Volume: 361

Issue: Unknown

Page Range: 483-495

Description:

In the past half century of fuzzy systems they were used to solve a wide range of complex problems, and the field of recommendation is no exception. The mathematical properties and the ability to efficiently process uncertain data enable fuzzy systems to face the common challenges in recommender systems. The main contribution of this paper is to give a comprehensive literature overview of various fuzzy based approaches to the solving of common problems and tasks in recommendation systems. As a conclusion possible new areas of research are discussed.

Open Access: Yes

DOI: 10.1007/978-3-319-75408-6_37

Comparing the properties of meta-heuristic optimization techniques with various parameters on a fuzzy rule-based classifier

Alex Tormási L. T. Koczy

Publication Name: Studies in Fuzziness and Soft Computing

Publication Date: 2016-01-01

Volume: 342

Issue: Unknown

Page Range: 157-169

Description:

In this paper, the results of meta-heuristic optimization techniques with various parameter settings are presented. A formerly published Fuzzy-Based Recognizer (FUBAR): A fuzzy rule-based classification algorithm was used to analyze and evaluate the behavior of the used meta-heuristic optimization algorithms for rule-base optimization. Besides the reached accuracy, the execution time, the CPU load of the algorithms, and the effects of the shapes of the fuzzy membership functions in the initial rule-base are also investigated.

Open Access: Yes

DOI: 10.1007/978-3-319-32229-2_12

Meta-heuristic optimization of a fuzzy character recognizer

Alex Tormási L. T. Koczy

Publication Name: Studies in Fuzziness and Soft Computing

Publication Date: 2015-01-01

Volume: 326

Issue: Unknown

Page Range: 227-244

Description:

Meta-heuristic algorithms are well researched and widely used in optimization problems. There are several meta-heuristic optimization algorithms with various concepts and each has its own advantages and disadvantages. Still it is difficult to decide which method would fit the best to a given problem. In this study the optimization of a fuzzy rule-base from a classifier, more specifically fuzzy character recognizer is used as the reference problem and the aim of the research was to investigate the behavior of selected meta-heuristic optimization techniques in order to develop a multi meta-heuristic algorithm.

Open Access: Yes

DOI: 10.1007/978-3-319-19683-1_13

Fuzzy single-stroke character recognizer with various rectangle fuzzy grids

Alex Tormási L. T. Koczy

Publication Name: Studies in Computational Intelligence

Publication Date: 2014-02-03

Volume: 530

Issue: Unknown

Page Range: 145-159

Description:

In this chapter we introduce the results of a formerly published FUBAR character recognition method with various fuzzy grid parameters. The accuracy and efficiency of the handwritten single-stroke character recognition algorithm with different sized rectangle (N×M) fuzzy grids are investigated. The results are compared to other modified FUBAR algorithms and known commercial and academic recognition methods. Possible applications and further extensions are also discussed. This work is the extended and fully detailed version of a previously published abstract. © Springer International Publishing Switzerland 2014.

Open Access: Yes

DOI: 10.1007/978-3-319-03206-1_11

Improving the accuracy of a fuzzy-based single-stroke character recognizer by antecedent weighting

Alex Tormási L. T. Koczy

Publication Name: Studies in Fuzziness and Soft Computing

Publication Date: 2014-01-01

Volume: 317

Issue: Unknown

Page Range: 165-179

Description:

In this chapter we present an improved version of the fuzzy based single-stroke character recognizer introduced in previous works. The modified recognition method is able to reach higher accuracy in the character recognition without any significant effect on the computational complexity of the algorithm. Different fuzzy rule and antecedent weighting techniques were successfully used to improve the efficiency of fuzzy systems especially in classification problems. The altered recognizer reached 99.49 % average recognition rate with 26 different single-stroke symbols (based on Palm’s Graffiti alphabet) without learning userspecific parameters or modifying the rule-base. The new algorithm has the same computational complexity as the original system does.

Open Access: Yes

DOI: 10.1007/978-3-319-06323-2_11

Dynamic fuzzy rule weight optimization for a Fuzzy Based Single-Stroke Character Recognizer

Alex Tormási L. T. Koczy

Publication Name: Ines 2013 IEEE 17th International Conference on Intelligent Engineering Systems Proceedings

Publication Date: 2013-12-12

Volume: Unknown

Issue: Unknown

Page Range: 119-124

Description:

In this paper a dynamic fuzzy rule weighting method (DFW) combined with evolutionary optimization are presented for the formerly published Fuzzy Based Single-Stroke Character Recognizer (FUBAR) method. With the introduced rule weighting technique the consequent parts of the if...then... rules are calculated similarly to the original FUBAR method, but a dynamic fuzzy rule weight Wn([0,1]) described as a fuzzy set is applied to it in O n·1/Wn(On) form, where On is the output of the rule. The membership functions of DFW-s are determined by bacterial evolutionary algorithm. The paper compares the results of the proposed new algorithm with other (formerly published) FUBAR algorithms and also with other commercial and academic single-stroke recognizers in terms of recognition accuracy and computational resources needed. © 2013 IEEE.

Open Access: Yes

DOI: 10.1109/INES.2013.6632795

Fuzzy-based multi-stroke character recognizer

Alex Tormási L. T. Koczy

Publication Name: 2013 Federated Conference on Computer Science and Information Systems Fedcsis 2013

Publication Date: 2013-12-01

Volume: Unknown

Issue: Unknown

Page Range: 671-674

Description:

In this paper an extension for multi-stroke character recognition of FUzzy BAsed handwritten character Recognition (FUBAR) algorithm will be presented. First the basic concept of a single-stroke version will be overviewed; in the second part of the paper the new version of the algorithm with multi-stroke symbol support will be introduced, which deploy the same algorithm overviewed in the first part and use flat and hierarchical rule bases. © 2013 Polish Information Processing Society.

Open Access: Yes

DOI: DOI not available

Doctoral symposium on recent advances in information technology

Ferenc Lilik Piotr Kulczycki L. T. Koczy Piotr Andrzej Kowalski Dominika Golunska Szymon Lukasik Jaroslaw Arabas Alex Tormási Radko Mesiar Krisztián Balázs Krassimir T. Atanassov Modesto Castrillon-Santana Malgorzata Charytanowicz Thomas Corpetti Nicolas Courty Philippe Fournier-Viger David Gil Bao Gang Hu Carles Noguera I Clofent Milan Petrik Yujiu Yang Slawomir Zadrozny

Publication Name: 2013 Federated Conference on Computer Science and Information Systems Fedcsis 2013

Publication Date: 2013-12-01

Volume: Unknown

Issue: Unknown

Page Range: 637

Description:

No description provided

Open Access: Yes

DOI: DOI not available

Improved fuzzy-based single-stroke character recognizer

Alex Tormási L. T. Koczy

Publication Name: Proceedings of the 2013 Joint Ifsa World Congress and NAFIPS Annual Meeting Ifsa NAFIPS 2013

Publication Date: 2013-10-31

Volume: Unknown

Issue: Unknown

Page Range: 430-435

Description:

In this paper we present two modified and improved versions of the formerly published Fuzzy-Based Single-Stroke Character Recognizer (FUBAR) algorithm. After introducing the original method, the study investigates the effects of two different improvements of the designed algorithm. The first extension is the use of symbol-dependent fuzzy grids to extract symbol features; the second one is the use of rule weights in hierarchical rule-bases. The accuracy and efficiency of the extended FUBAR algorithms are compared to previous results. © 2013 IEEE.

Open Access: Yes

DOI: 10.1109/IFSA-NAFIPS.2013.6608439

Single-stroke character recognition with fuzzy method

J. Botzheim Alex Tormási

Publication Name: Studies in Computational Intelligence

Publication Date: 2013-01-01

Volume: 417

Issue: Unknown

Page Range: 27-46

Description:

In this paper an on-line single-stroke recognition method based on fuzzy logic is introduced. Each of the characters is defined by only one nine dimensional fuzzy rule. In addition to the low resource requirement the solution is able to satisfy many of the user's current demands in handwriting recognizers, like speed and learning. Eight of the nine features are extracted using a four-by-four grid. For the learning phase we designed a new punish/reward bacterial evolutionary algorithm which tunes the character parameters represented by fuzzy sets. © 2013 Springer-Verlag Berlin Heidelberg.

Open Access: Yes

DOI: 10.1007/978-3-642-28959-0_2

Improving the efficiency of a fuzzy-based single-stroke character recognizer with hierarchical rule-base

Alex Tormási L. T. Koczy

Publication Name: Cinti 2012 13th IEEE International Symposium on Computational Intelligence and Informatics Proceedings

Publication Date: 2012-12-01

Volume: Unknown

Issue: Unknown

Page Range: 421-426

Description:

In this paper we present an improved version of the fuzzy based single-stroke character recognizer introduced in previous works. The modified recognition method is able to reach an acceptable accuracy in the character recognition with a significant decrease on the computational complexity of the algorithm. Different hierarchical rule-base techniques were successfully used to improve the efficiency of fuzzy systems. The altered recognizer reached 98.82% average recognition rate with 26 different single-stroke symbols (based on Palm's Graffiti alphabet) without learning user-specific parameters or modifying the rule-base during the tests. The new algorithm has a small decrease in the recognition rate compared to the accuracy of the original systems but the new method has less computational price than the original system does. © 2012 IEEE.

Open Access: Yes

DOI: 10.1109/CINTI.2012.6496803

Comparing the efficiency of a fuzzy single-stroke character recognizer with various parameter values

Alex Tormási L. T. Koczy

Publication Name: Communications in Computer and Information Science

Publication Date: 2012-11-02

Volume: 297 CCIS

Issue: PART 1

Page Range: 260-269

Description:

In this paper the results of a study on the accuracy of a fuzzy logic-based single-stroke character recognizer are presented by refining various parameter values, such as resolution of the fuzzy grid and the minimum distance between sampled points. The symbol set is a modified version of Palm's Graffiti single-stroke alphabet and it contains 26 different symbols. Each symbol is represented by a single fuzzy rule. The rule base was determined by a subset of the collected samples. 99.4% recognition rate has been achieved with the initial rule base, without training. With the revised parameter values the accuracy is close or even slightly beyond the results of other academic or commercial systems. © 2012 Springer-Verlag Berlin Heidelberg.

Open Access: Yes

DOI: 10.1007/978-3-642-31709-5_27