KINDAI UNIVERSITY


*A space between the first name and last name, please enter

OHTANI Takashi

Profile

FacultyDepartment of Informatics
PositionLecturer
Degree
Commentator Guidehttps://www.kindai.ac.jp/meikan/463-ootani-takashi.html
URL
Mail
Last Updated :2020/09/03

Research Activities

Research Areas

  • Informatics, Sensitivity (kansei) informatics
  • Informatics, Soft computing

Research Interests

  • Newrofuzzy Adaptive Modelling

Published Papers

  • A Design Method To Extend Stations To The Existing Flexible Mixed-Product Line, International Journal of Business Research, International Journal of Business Research, 10(4), 45 - 51, Oct. 2010
  • A Redesign Method for Flexible Mixed-product Lines Adapted to the Change of the Production Ratio, International Journal of Business Research, International Journal of Business Research, 9(5), 45 - 53, Oct. 2009
  • A DESIGN METHOD FOR FLEXIBLE MIXED-PRODUCT LINES AVAILABLE IN VARIOUS CONDITIONS, Aritoshi Kimura, Tatsuo Matsutomi, Takashi Ohtani, ICIM2012: PROCEEDINGS OF THE ELEVENTH INTERNATIONAL CONFERENCE ON INDUSTRIAL MANAGEMENT, ICIM2012: PROCEEDINGS OF THE ELEVENTH INTERNATIONAL CONFERENCE ON INDUSTRIAL MANAGEMENT, 109 - 114, 2012
    Summary:Product lines are usually designed in various conditions. If requirements of the production change, the product line should be adjusted, to the suitable one. There are many types of optimization method, such as addition or reduction of stations in the production line in succession to the structure of the existing line. In this paper, we propose a design method for flexible mixed-product lines in various conditions. We have proposed the Fissiparous Algorithm and already shown the effectiveness by designing flexible mixed-product lines. We developed the Fissiparous Algorithm by new operations and used it to design flexible mixed-product lines with various product conditions. The validity of this design method using the Fissiparous Algorithm is shown through solving several design problems of flexible mixed-product lines.

Conference Activities & Talks

  • Advanced Fissiparous Algorithm for Designing Flexible Mixed-Product Lines, The nines International Conference on Industrial Management,   2008 09 , The nines International Conference on Industrial Management

Misc

  • Automatic variable selection in radial basis functions network, T Ohtani, ISIM'2000: PROCEEDINGS OF THE FIFTH CHINA-JAPAN INTERNATIONAL SYMPOSIUM ON INDUSTRIAL MANAGEMENT, 409, 414,   2000
    Summary:The radial basis functions (RBF) network is a technique for interpolating data in high dimensional spaces and regarded as one of neural networks. It is important, but difficult to select the necessary input variables to the network among many variables because we do not have knowledge about the target system which is identified by the network. The computation to find the optimal combination of variables is intense. In this paper, we consider the method to automatically select the optimal combination of variables in one learning iteration. The affine scaling, interior point method of Dikin is known as a simple and effective algorithm of linear programming. And, the gradient projection method of Rosen is known as an effective algorithm of nonlinear programming. We propose a combined algorithm of the affine scaling interior point method and the gradient projection method for the automatic selection of the optimal combination of variables. This method simultaneously enables the learning of model parameters and the selection of variables. Numerical examples show the validity of the method.
  • Automatic Variable Selection in Radial Basis Functions Network and Its Application to Neurofuzzy GMDH, Proc. 4th Int. Conf. Knowledge-Based Intelligent Engineering Systems & Allied Technologies, 2, 840, 843,   2000
  • Automatic Variable Selection in Neurofuzzy GMDH with Successive Determination of Variables, Proc. 15th Int. Conf. Production Research, 1, 513, 516,   1999
  • Automatic Variable Selection in Neurofuzzy GMDH with Successive Determination of Variables, The Trans Institute of Electronic, Information and Communication Engineers D-(]G0002[), J28-D-II, 9, 1492, 1499,   1999
  • Evaluation of Senior-Simulation by Corresponding Analysis with Real Value Power Exponent Norm, Journal of Japan Industrial Management Association, 49, 6.339-343,   1999 , 10.11221/jima.49.339
  • Neurofuzzy cognitive maps with adaptive B-splines, T Ohtani, H Ichihashi, T Miyoshi, INFORMATION SCIENCES, 110, 1-2, 81, 102,   1998 09 , 10.1016/S0020-0255(97)10078-0
    Summary:Cognitive maps and time space maps make use of multidimensional scaling (MDS) techniques to analyze data relating to spatial and environmental preferences and perception. Perceptual configuration of the points is represented by a cognitive map with surface feature interpolation. In this paper we propose three procedures of MDS using Neurofuzzy adaptive modelling with B-splines for surface feature interpolation. The procedures are based on: (1) Hayashi's quantifying method of paired comparisons, (2) Torgerson's metrical MDS procedure and (3) Gradient descent method which is the basic learning method of adaptive systems such as the artificial neural networks and neurofuzzy modelling. In numerical examples, the resultant maps are compared and an application to sociometry analysis is presented. (C) 1998 Elsevier Science Inc. All rights reserved.
  • Orthogonal and successive projection methods for the learning of neurofuzzy GMDH, T Ohtani, H Ichihashi, T Miyoshi, K Nagasaka, INFORMATION SCIENCES, 110, 1-2, 5, 24,   1998 09 , 10.1016/S0020-0255(97)10082-2
    Summary:A neural GMDH (group method of data handling) family of modelling algorithm emulates the self-organizing activity of the central nervous system, and discovers the structure (functional form) of empirical models that include many input variables. A generalized successive projection method is developed for the accelerated learning algorithm of the GMDH type model whose partial descriptions are represented by the radial basis functions network, (1) For the learning of partial descriptions of the perceptron type GMDH, a combined algorithm of the successive projection method and the orthogonal projection method is developed. (2) For the learning of the network type GMDH, a successive projection method is derived as the solution of an optimization problem in which the Minkowski norm of distance travelled (step size) is minimized. Their performances are compared with the instantaneous learning algorithms such as the least mean square. Several examples show the validity of the methods. (C) 1998 Elsevier Science Inc. All rights reserved.
  • A pointing Device Using Coordinate Transformation by Neurofuzzy GMDH, Journal of Japan Society for Fuzzy Theory and Systems, 10, 2, 322, 329,   1998 , 10.3156/jfuzzy.10.2_142
  • Selection of Number of Layers with Distorter in Neurofuzzy GMDH, The Transaction of the Institute of Electronics, Information and Communication Engineers D-(]G0002[), J81-D-(]G0002[), 2, 396, 403,   1998
  • Structural learning with M-apoptosis in neurofuzzy GMDH, T Ohtani, H Ichihashi, T Miyoshi, K Nagasaka, 1998 IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS AT THE IEEE WORLD CONGRESS ON COMPUTATIONAL INTELLIGENCE - PROCEEDINGS, VOL 1-2, 11, 5, 1265, 1270,   1998 , 10.5687/iscie.11.251
    Summary:There have been many studies of mathematical models of neural networks. However there always arises a problem of determining their optimal structures because of the lark of prior information. Apoptosis is the mechanism responsible for the physiological deletion of cells and appears to be intrinsically programmed. We propose a procedure named M-apoptosis for the structure clarification of Neurofuzzy GMDH model whose partial descriptions are represented by the Radial Basis functions network. The proposed method prunes unnecessary links and units hom the larger network to identify, still more to clarify the network structure by minimizing the Minkowski norm of the derivatives of the partial descriptions. The method is validated in the numerical examples of function approximation and the classification of Fisher's Iris data.
  • Neurofuzzy Cognitive Maps with Adaptive B-splines, Journal of Japan Society for Fuzzy and Systems, 9, 5, 731, 746,   1997 , 10.3156/jfuzzy.9.5_737
  • A Pointing Device Using Hand and Fingers Equipped with a Multi-color Tracker (共著), Biomedical Fuzzy and Human Sciences, 3, 1, 11, 20,   1997
  • Selection of the Number of Layers in Neuro-Fuzzy GMDH, Journal of Japan Industrial Management Association, 9, 4, 125, 132,   1997
  • Succsessive Projection Method for Learning of Neurofuzzy GMDH, Journal of Japan Society for Fuzzy Theory and Systems, 9, 4, 472, 484,   1997 , 10.3156/jfuzzy.9.4_472
  • Function Approximation by Neurofuzzy GMDH with Error Backpropagation Learning, Journal of Japan Industrial Management Association, 47, 6, 384, 392,   1997 , 10.11221/jima.47.384

Research Grants & Projects

  • Study on GMDH Type Adaptive Learning Networks