| Analysis on the Comparison Exponential Smoothing and Neural Network in Forecasting the Trend of Toddler Nutritions in Community Health Centre |
| Pages: 207-215 (9) | [Full Text] PDF (981 KB) |
| AB Santoso, D Manongga, I Sembiring |
| Fakultas Teknologi Informasi, Universitas Kristen Satya Wacana, UKSW Jl. Diponegoro 52-60, Salatiga 50711, Indonesia |
|
Abstract - In Indonesia, community health centres (a.k.a Puskesmas) provide integrated services and consultations for the communities, including toddler care services. There is an assumption that the increase of toddlers nutrition status is influenced by toddlers age and parents economic status. In this study, the exponential smoothing and the neural network methods were used to forecast toddlers nutrition status. The forecastings were then used to test the assumption whether toddlers nutrition status may be influenced by parents economic status, education, and care status. The forecasting by exponential smoothing with Eviews method and Neural Network Backpropagation method with Matlab were analyzed and compared to determine which forecasting was best for the next three months based on the pattern of the previous trend data. The results of analysis are used to facilitate and assist community health centre officers in forecasting and evaluating the increase in toddlers nutrition status. |
| Index Terms - Exponential Smoothing, Neural Network Backpropagation, Forecasting, Eviews, Matlab |
|
C itation - AB Santoso, D Manongga, I Sembiring. "Analysis on the Comparison Exponential Smoothing and Neural Network in Forecasting the Trend of Toddler Nutritions in Community Health Centre." International Journal of Computer Science and Software Engineering 6, no. 10 (2017): 207-215. |
| Image Processing Based Analysis of the Compressive Strength for the Stones Used In Historical Masonry Structures |
| Pages: 216-222 (7) | [Full Text] PDF (758 KB) |
| SG Ozkaya, M Baygin, MA Ozdemir, I Kazaz |
| Civil Engineering, Computer Engineering, Ardahan University, Ardahan, 75000, TurkeyCivil Engineering, Igdir University, Igdir, 76000, TurkeyCivil Engineering, Erzurum Technical University, Erzurum, 25000, Turkey |
|
Abstract - Determining the mechanical properties of materials used in masonry structures, which is one of the most important issues in civil engineering, is a field that requires very laborious and intensive laboratory work. As a result of these studies, basic mechanical properties such as compressive strength and tensile strength of structures can be obtained. These parameters, especially used in the restoration process of these structures, are the basic arguments for determining the real behavior of the structure. With this study, it was provided to examine the stones used in the historical masonry structures with computer vision technology. For this purpose, stones having different qualities taken from stone quarries are primarily taken through a camera. Subsequently, the image of each sample is transferred to the computer environment and the properties of these samples are obtained by image processing. After all these tests, these samples are tested in the laboratory and compressive strength are measured. As a result, the data obtained from the laboratory environment and the results obtained by image processing are compared and the calibration of this proposed image processing based analysis method is provided. With this new approach, which is a recommended image processing base, it is possible to perform experimental applications in the laboratory environment in computer environment. Again due to these studies very difficult and lengthy studies have been shortened to a considerable extent and the specific features of the images taken through a basic camera have been calculated very quickly in the computer environment. During the study, approximately 100 pieces of stone samples were provided and 80% of these samples were divided into two groups: training and 20% test data. Image processing based analysis method was applied to both groups and training data was used for calibration of the system. In the proposed approach, both image processing based analysis approach and laboratory crushing experiment were performed for each sample. As a result of all these tests, the proposed approach has been able to analyze the pressure strength on the stones with a margin of error of about 2%. |
| Index Terms - Image Processing, Artificial Neural Network, Compressive Strength, Stone |
|
C itation - SG Ozkaya, M Baygin, MA Ozdemir, I Kazaz. "Image Processing Based Analysis of the Compressive Strength for the Stones Used In Historical Masonry Structures." International Journal of Computer Science and Software Engineering 6, no. 10 (2017): 216-222. |
| Modeling Composition of UML Profiles with Alloy |
| Pages: 223-232 (10) | [Full Text] PDF (660 KB) |
| K Farias, T Oliveira, LJ Goncales, V Bischoff |
| Post Graduate Program in Applied Computing (PPGCA), University of Vale do Rio dos Sinos (UNISINOS), Sao Leopoldo, RS 93.022-750, BrazilComputation and Systems engineering Program (PESC/COPPE), Federal University of Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brazil |
|
Abstract - Global software development teams can collaboratively create Unified Modeling Language (UML) profiles, the primary mechanism for defining domain-specific variants on top of the UML. Usually, parts of UML profiles are separately elaborated to speed up the UML tailoring process, but at sometimes the parts built in parallel need to be brought together to construct a full UML profile. Although many model composition techniques have been proposed in the last decades, no one deals with issues required to combine UML profiles, e.g., matching and integration of UML stereotypes. Consequently, little is known about how to support the composition of UML profiles. Even worse, academia and industry have overlooked the elaboration of composition methods to support the integration of UML profiles, as well as the formal representation of such methods. This study, therefore, presents a composition mechanism, as well as introduces a lightweight UML extension to support the specification of composition relationships between UML profiles. The semantics of the extension and mechanism proposed was carefully represented in Alloy, a formal modeling language based on first-order logic. Then, we used the Alloy Analyzer to check the specification generated in Alloy for some specific algebraic properties, including idempotency, uniqueness, commutativity, and associativity. |
| Index Terms - Alloy, Model Composition, UML, UML profile |
|
C itation - K Farias, T Oliveira, LJ Goncales, V Bischoff. "Modeling Composition of UML Profiles with Alloy." International Journal of Computer Science and Software Engineering 6, no. 10 (2017): 223-232. |