Time filter

Source Type

Basuki S.,University of Muhammadiyah Malang | Purwarianti A.,Bandung Institute of Technology
International Journal on Electrical Engineering and Informatics | Year: 2016

This research has proposed a method to decompose complex factoid question into several independent questions. The method comprises four stages: (1) classifying input question into several categories such as sub-question, coordination, exemplification, or double question, (2) generating all possible question boundary candidates, (3) selecting the best question boundary, and (4) performing the question decomposition rule using the best question boundary. This study compared several machine learning algorithms in the first stage (complex factoid question classification) and third stage (question decomposition boundary selection). The features used in the classification are specific word lists with its related information including the syntactic features of POS (Part of Speech) tag. For the experiments, we annotated 916 sentences for training data and 226 sentences for testing data. The perplexity of the annotated corpus achieved 1.000586 with 307 Out of Vocabulary (OOV). The complex factoid question classification accuracy reached 93.8% with Random Forest algorithm. The question decomposition boundary selection accuracy achieved 93.80% for sub-question (using Random Forest algorithm), 86.11% for double question (using Random Forest algorithm), 88.23% for coordination (using SMO), and 60.87% for exemplification (using kNN, NB, and RF). A revision rule was provided for the question decomposition boundary selection that improved the accuracy into 97.22% for double question, 94.11% for coordination, and 65.21% for exemplification. © 2016, School of Electrical Engineering and Informatics. All rights reserved. Source

Faruq A.,University of Muhammadiyah Malang | Abdullah S.S.B.,University of Technology Malaysia | Shah M.F.N.,University of Technology Malaysia
Telkomnika | Year: 2011

Underwater environment poses a difficult challenge for autonomous underwater navigation. A standard problem of underwater vehicles is to maintain it position at a certain depth in order to perform desired operations. An effective controller is required for this purpose and hence the design of a depth controller for an unmanned underwater vehicle is described in this paper. The control algorithm is simulated by using the marine guidance navigation and control simulator. The project shows a radial basis function metamodel can be used to tune the scaling factors of a fuzzy logic controller. By using offline optimization approach, a comparison between genetic algorithm and metamodeling has been done to minimize the integral square error between the set point and the measured depth of the underwater vehicle. The results showed that it is possible to obtain a reasonably good error using metamodeling approach in much a shorter time compared to the genetic algorithm approach. © 2011 Universitas Ahmad Dahlan. Source

Sukorinia H.,Kasetsart University | Sukorinia H.,University of Muhammadiyah Malang | Sangchote S.,Kasetsart University | Khewkhomc N.,Kasetsart University
Postharvest Biology and Technology | Year: 2013

The use of bio-fungicides and a plant extracts to control postharvest disease was investigated as an alternative to chemical control. The combination of a promising plant extract and yeast were selected through in vitro and in vivo techniques. A combination of Candida utilis TISTR 5001 and Eugenia caryophyllata crude extract was the best combination to attain a reduction in disease incidence and disease severity of Penicillium digitatum on citrus fruit. Colonization was the lowest on fruit treated with the combination of E. caryophylata crude extract and C. utilis TISTR 5001, and survival of C. utilis TISTR 5001 was the highest. The combination of E. caryophylata crude extract and C. utilis TISTR 5001 significantly reduced the natural development of green mold of citrus fruit, and had no effect to fruit quality. The active compound of E. caryophylata was found to be eugenol, based on HPLC and NMR (1H and 13C). Hence, the results indicate that a combination of plant extracts and yeasts posses antifungal activity that can be exploited as an ideal treatment for future plant disease management. © 2013 Elsevier B.V. Source

Minarno A.E.,University of Muhammadiyah Malang | Suciati N.,Sepuluh Nopember Institute of Technology
Telkomnika (Telecommunication Computing Electronics and Control) | Year: 2014

Study in batik image retrieval is still challenging today. One of the methods for this problem is using a Color Difference Histogram (CDH), which is based on the difference of color features and edge orientation features. However, CDH is only utilising local features instead of global features; consequently it cannot represent images globally. We suggest that by adding global features for batik image retrieval, precision will increase. Therefore, in this study, we combine the use of modified CDH to define local features and the use of Grey Level Co-occurrence Matrix (GLCM) to define global features. The modified CDH is performed by changing the size of image quantisation, so it can reduce the number of features. Features that are detected by GLCM are energy, entropy, contrast and correlation. In this study, we use 300 batik images which consist of 50 classes and six images in each class. The experiment result shows that the proposed method is able to raise 96.5% of the precision rate which is 3.5% higher than the use of CDH only. The proposed method is extracting a smaller number of features; however it performs better for batik image retrieval. This indicates that the use of GLCM is effective combined with CDH. Source

Minarno A.E.,University of Muhammadiyah Malang | Suciati N.,Sepuluh Nopember Institute of Technology
Journal of Theoretical and Applied Information Technology | Year: 2014

One of many method for image retrieval is Multi Texton Histogram (MTH) that incorporated feature extraction technique. Though the MTH is able to represent the image very well, it’s still has weaknesses. First, the MTH is only using local features to represent image. Second, in the process of pixel pair detection using texton, there is information missing that caused image representation may degrade. This study proposes a new method in order to extract image features for the image retrieval system. The proposed method is named Multi Texton Co-Occurrence Descriptor (MTCD). The MTCD is extracting color, texture and shape features simultaneously using texton, and then calculating image representation globally using Gray Level Co-occurrence Matrix (GLCM). This study used 300 Batik images and 15000 Corel images as datasets. Image similarity is calculated using Canberra and MTCD performance is measured using precision and recall. Our experiments show that by adding 2 new textons and GLCM, the precision rate is increased by 2.86% for Batik dataset, by 3.40%for Corel 5,000 and by 3.06% for Corel 10,000. We conclude that MTCD performance is superior than MTH. © 2005 - 2014 JATIT & LLS. All rights reserved. Source

Discover hidden collaborations