Winarko, Edi
Computer Science And Electronics Department, Faculty Of Mathematics And Natural Sciences Universitas Gadjah Mada, Yogyakarta

Published : 12 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search
Journal : International Journal of Electrical and Computer Engineering

Ontology-based Why-Question Analysis Using Lexico-Syntactic Patterns A.A.I.N. Eka Karyawati; Edi Winarko; Azhari Azhari; Agus Harjoko
International Journal of Electrical and Computer Engineering (IJECE) Vol 5, No 2: April 2015
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (186.282 KB) | DOI: 10.11591/ijece.v5i2.pp318-332

Abstract

This research focuses on developing a method to analyze why-questions.  Some previous researches on the why-question analysis usually used the morphological and the syntactical approach without considering the expected answer types. Moreover, they rarely involved domain ontology to capture the semantic or conceptualization of the content. Consequently, some semantic mismatches occurred and then resulting not appropriate answers. The proposed method considers the expected answer types and involves domain ontology. It adapts the simple, the bag-of-words like model, by using semantic entities (i.e., concepts/entities and relations) instead of words to represent a query. The proposed method expands the question by adding the additional semantic entities got by executing the constructed SPARQL query of the why-question over the domain ontology. The major contribution of this research is in developing an ontology-based why-question analysis method by considering the expected answer types. Some experiments have been conducted to evaluate each phase of the proposed method. The results show good performance for all performance measures used (i.e., precision, recall, undergeneration, and overgeneration). Furthermore, comparison against two baseline methods, the keyword-based ones (i.e., the term-based and the phrase-based method), shows that the proposed method obtained better performance results in terms of MRR and P@10 values.
Survey: Models and Prototypes of Schema Matching Edhy Sutanta; Retantyo Wardoyo; Khabib Mustofa; Edi Winarko
International Journal of Electrical and Computer Engineering (IJECE) Vol 6, No 3: June 2016
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (167.178 KB) | DOI: 10.11591/ijece.v6i3.pp1011-1022

Abstract

Schema matching is critical problem within many applications to integration of data/information, to achieve interoperability, and other cases caused by schematic heterogeneity. Schema matching evolved from manual way on a specific domain, leading to a new models and methods that are semi-automatic and more general, so it is able to effectively direct the user within generate a mapping among elements of two the schema or ontologies better. This paper is a summary of literature review on models and prototypes on schema matching within the last 25 years to describe the progress of and research chalenge and opportunities on a new models, methods, and/or prototypes.
Face Recognition Based on Symmetrical Half-Join Method using Stereo Vision Camera Edy Winarno; Agus Harjoko; Aniati Murni Arymurthy; Edi Winarko
International Journal of Electrical and Computer Engineering (IJECE) Vol 6, No 6: December 2016
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (435.818 KB) | DOI: 10.11591/ijece.v6i6.pp2818-2827

Abstract

The main problem in face recognition system based on half-face pattern is how to anticipate poses and illuminance variations to improve recognition rate. To solve this problem, we can use two lenses on stereo vision camera in face recognition system. Stereo vision camera has left and right lenses that can be used to produce a 2D image of each lens. Stereo vision camera in face recognition has capability to produce two of 2D face images with a different angle. Both angle of the face image will produce a detailed image of the face and better lighting levels on each of the left and right lenses. In this study, we proposed a face recognition technique, using 2 lens on a stereo vision camera namely symmetrical half-join. Symmetrical half-join is a method of normalizing the image of the face detection on each of the left and right lenses in stereo vision camera, then cropping and merging at each image. Tests on face recognition rate based on the variety of poses and variations in illumination shows that the symmetrical half-join method is able to provide a high accuracy of face recognition and can anticipate variations in given pose and illumination variations. The proposed model is able to produce 86% -97% recognition rate on a variety of poses and variations in angles between 0 °- 22.5 °. The variation of illuminance measured using a lux meter can result in 90% -100% recognition rate for the category of at least dim lighting levels (above 10 lux).
A Hybrid Model Schema Matching Using Constraint-Based and Instance-Based Edhy Sutanta; Retantyo Wardoyo; Khabib Mustofa; Edi Winarko
International Journal of Electrical and Computer Engineering (IJECE) Vol 6, No 3: June 2016
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (238.058 KB) | DOI: 10.11591/ijece.v6i3.pp1048-1058

Abstract

Schema matching is an important process in the Enterprise Information Integration (EII) which is at the level of the back end to solve the problems due to the schematic heterogeneity. This paper is a summary of preliminary result work of the model development stage as part of research on the development of models and prototype of hybrid schema matching that combines two methods, namely constraint-based and instance-based. The discussion includes a general description of the proposed models and the development of models, start from requirement analysis, data type conversion, matching mechanism, database support, constraints and instance extraction, matching and compute the similarity, preliminary result, user verification, verified result, dataset for testing, as well as the performance measurement. Based on result experiment on 36 datasets of heterogeneous RDBMS, it obtained the highest P value is 100.00% while the lowest is 71.43%; The highest R value is 100.00% while the lowest is 75.00%; and F-Measure highest value is 100.00% while the lowest is 81.48%. Unsuccessful matching on the model still happens, including use of an id attribute with data type as autoincrement; using codes that are defined in the same way but different meanings; and if encountered in common instance with the same definition but different meaning.
Content-based product image retrieval using squared-hinge loss trained convolutional neural networks Arif Rahman; Edi Winarko; Khabis Mustofa
International Journal of Electrical and Computer Engineering (IJECE) Vol 13, No 5: October 2023
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijece.v13i5.pp5804-5812

Abstract

Convolutional neural networks (CNN) have proven to be highly effective in large-scale object detection and image classification, as well as in serving as feature extractors for content-based image retrieval. While CNN models are typically trained with category label supervision and softmax loss for product image retrieval, we propose a different approach for feature extraction using the squared-hinge loss, an alternative multiclass classification loss function. First, transfer learning is performed on a pre-trained model, followed by fine-tuning the model. Then, image features are extracted based on the fine-tuned model and indexed using the nearest-neighbor indexing technique. Experiments are conducted on VGG19, InceptionV3, MobileNetV2, and ResNet18 CNN models. The model training results indicate that training the models with squared-hinge loss reduces the loss values in each epoch and reaches stability in less epoch than softmax loss. Retrieval results show that using features from squared-hinge trained models improves the retrieval accuracy by up to 3.7% compared to features from softmax-trained models. Moreover, the squared-hinge trained MobileNetV2 features outperformed others, while the ResNet18 feature gives the advantage of having the lowest dimensionality with competitive accuracy.