Analysis of Segmentation Parameters Effect towards Parallel Processing Time on Fuzzy C Means Algorithm

2018 ◽  
Vol 1 (4) ◽  
pp. 99
Author(s):  
Cepi Ramdani ◽  
Indah Soesanti ◽  
Sunu Wibirama
Keyword(s):  
Parallel Processing ◽  
Processing Time ◽  
Color Image ◽  
Clustering Algorithms ◽  
Fuzzy C Means ◽  
Sequential Processing ◽  
Processing Times ◽  
Parameter Values ◽  
Fuzzy C Means Algorithm ◽  
Segmentation Parameters

Fuzzy C Means algorithm or FCM is one of many clustering algorithms that has better accuracy to solve problems related to segmentation. Its application is almost in every aspects of life and many disciplines of science. However, this algorithm has some shortcomings, one of them is the large amount of processing time consumption. This research conducted mainly to do an analysis about the effect of segmentation parameters towards processing time in sequential and parallel. The other goal is to reduce the processing time of segmentation process using parallel approach. Parallel processing applied on Nvidia GeForce GT540M GPU using CUDA v8.0 framework. The experiment conducted on natural RGB color image sized 256x256 and 512x512. The settings of segmentation parameter values were done as follows, weight in range (2-3), number of iteration (50-150), number of cluster (2-8), and error tolerance or epsilon (0.1 – 1e-06). The results obtained by this research as follows, parallel processing time is faster 4.5 times than sequential time with similarity level of image segmentations generated both of processing types is 100%. The influence of segmentation parameter values towards processing times in sequential and parallel can be concluded as follows, the greater value of weight parameter then the sequential processing time becomes short, however it has no effects on parallel processing time. For iteration and cluster parameters, the greater their values will make processing time consuming in sequential and parallel become large. Meanwhile the epsilon parameter has no effect or has an unpredictable tendency on both of processing time.

Construct Knowledge Structure of Linear Algebra

2011 ◽  
Vol 211-212 ◽  
pp. 793-797
Author(s):  
Chin Chun Chen ◽  
Yuan Horng Lin ◽  
Jeng Ming Yih ◽  
Sue Fen Huang
Keyword(s):  
Knowledge Management ◽  
Linear Algebra ◽  
Fuzzy Clustering ◽  
Mahalanobis Distance ◽  
Clustering Algorithms ◽  
Knowledge Structure ◽  
Interpretive Structural Modeling ◽  
Cognitive Characteristics ◽  
Fuzzy C Means ◽  
Fuzzy C Means Algorithm

Apply interpretive structural modeling to construct knowledge structure of linear algebra. New fuzzy clustering algorithms improved fuzzy c-means algorithm based on Mahalanobis distance has better performance than fuzzy c-means algorithm. Each cluster of data can easily describe features of knowledge structures individually. The results show that there are six clusters and each cluster has its own cognitive characteristics. The methodology can improve knowledge management in classroom more feasible.

Management of Abstract Algebra Concepts Based on Knowledge Structure

2013 ◽  
Vol 284-287 ◽  
pp. 3537-3542
Author(s):  
Chin Chun Chen ◽  
Yuan Horng Lin ◽  
Jeng Ming Yih
Keyword(s):  
Mahalanobis Distance ◽  
Clustering Algorithm ◽  
Fuzzy Theory ◽  
Clustering Algorithms ◽  
Knowledge Structures ◽  
Interpretive Structural Modeling ◽  
Fuzzy C Means ◽  
Integrated Method ◽  
Combined Algorithm ◽  
Fuzzy C Means Algorithm

Knowledge Management of Mathematics Concepts was essential in educational environment. The purpose of this study is to provide an integrated method of fuzzy theory basis for individualized concept structure analysis. This method integrates Fuzzy Logic Model of Perception (FLMP) and Interpretive Structural Modeling (ISM). The combined algorithm could analyze individualized concepts structure based on the comparisons with concept structure of expert. Fuzzy clustering algorithms are based on Euclidean distance function, which can only be used to detect spherical structural clusters. A Fuzzy C-Means algorithm based on Mahalanobis distance (FCM-M) was proposed to improve those limitations of GG and GK algorithms, but it is not stable enough when some of its covariance matrices are not equal. A new improved Fuzzy C-Means algorithm based on a Normalized Mahalanobis distance (FCM-NM) is proposed. Use the best performance of clustering Algorithm FCM-NM in data analysis and interpretation. Each cluster of data can easily describe features of knowledge structures. Manage the knowledge structures of Mathematics Concepts to construct the model of features in the pattern recognition completely. This procedure will also useful for cognition diagnosis. To sum up, this integrated algorithm could improve the assessment methodology of cognition diagnosis and manage the knowledge structures of Mathematics Concepts easily.

scholarly journals Implementation of Parallel Processing on Multi-Object Recognition System Software

2018 ◽  
pp. 124
Author(s):  
Midriem Mirdanies
Keyword(s):  
Object Recognition ◽  
Parallel Processing ◽  
Processing Time ◽  
Application Programming Interface ◽  
Recognition System ◽  
Scale Invariant ◽  
Development Environment ◽  
Sequential Processing ◽  
Speeded Up Robust Features ◽  
Recognition Software

Multi-object recognition software on Remote Controlled Weapon Station (RCWS) had been implemented in previous paper using Scale Invariant Feature Transform (SIFT) and Speeded Up Robust Features (SURF) methods, but the processing time in one cycle is quite slow so it is need to be optimized using parallel processing. In this paper, implementation of parallel processing on multi-object recognition software has been done on a multicore processor. The Openmp Application Programming Interface (API), C programming language, and Visual studio Integrated Development Environment (IDE) is used to implement the parallel processing in this paper. The parallel processing was implemented in the for loop of the matching process between the capturing object from the camera and the database under two conditions, i.e., the original of the for loop syntax and after optimization of the for loop syntax. Experiments have been done on the core processor i7-4790 @ 3.60Ghz, 8 GB DDR3 of memory, windows 8.1 os using two, four, six, and eight cores to recognize one, two, three and four objects at once using SIFT and SURF methods. Based on the experiments, it was found that the processing time in parallel is faster than sequential process, where the fastest of the processing time is obtained after optimization in the loop syntax, with the processing time in recognizing one to four objects using SIFT method is 927.13 ms (8 core), 1019.31 ms (6 core), 1190.72 ms (8 core), and 1283.05 ms (4 core), where the sequential processing time in recognizing one to four objects is 1067.35 ms, 1164.78 ms, 1352.93 ms, and 1497.35 ms, while the processing time in recognizing one to four objects using SURF method is 1157.13 ms (8 core), 1517.83 ms (6 core), 1572.14 ms (4 core), dan 1472.64 ms (6 core), where the sequential processing time in recognizing one to four objects is 5635.99 ms, 6268.47 ms, 3256.63 ms, dan 3883.78 ms.

Comparative study on textual data set using fuzzy clustering algorithms

Kybernetes ◽  
2016 ◽  
Vol 45 (8) ◽  
pp. 1232-1242 ◽  
Author(s):  
Rjiba Sadika ◽  
Moez Soltani ◽  
Saloua Benammou
Keyword(s):  
Comparative Study ◽  
Clustering Algorithms ◽  
Fuzzy Model ◽  
Data Sets ◽  
Data Set ◽  
Content Type ◽  
Fuzzy C Means ◽  
Textual Data ◽  
Tunisian Revolution ◽  
Fuzzy C Means Algorithm

Purpose The purpose of this paper is to apply the Takagi-Sugeno (T-S) fuzzy model techniques in order to treat and classify textual data sets with and without noise. A comparative study is done in order to select the most accurate T-S algorithm in the textual data sets. Design/methodology/approach From a survey about what has been termed the “Tunisian Revolution,” the authors collect a textual data set from a questionnaire targeted at students. Five clustering algorithms are mainly applied: the Gath-Geva (G-G) algorithm, the modified G-G algorithm, the fuzzy c-means algorithm and the kernel fuzzy c-means algorithm. The authors examine the performances of the four clustering algorithms and select the most reliable one to cluster textual data. Findings The proposed methodology was to cluster textual data based on the T-S fuzzy model. On one hand, the results obtained using the T-S models are in the form of numerical relationships between selected keywords and the rest of words constituting a text. Consequently, it allows the authors to interpret these results not only qualitatively but also quantitatively. On the other hand, the proposed method is applied for clustering text taking into account the noise. Originality/value The originality comes from the fact that the authors validate some economical results based on textual data, even if they have not been written by experts in the linguistic fields. In addition, the results obtained in this study are easy and simple to interpret by the analysts.

Sign in / Sign up

Export Citation Format

Share Document