• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.339-348
• /
• 2021

In recent software defect prediction research, defect prediction between cross projects and cross-version projects are actively studied. Cross-version defect prediction studies assume WP(Within-Project) so far. However, in the CV(Cross-Version) environment, the previous work does not consider the distribution difference between project versions is important. In this study, we propose an automated Bayesian optimization framework that considers distribution differences between different versions. Through this, it automatically selects whether to perform transfer learning according to the difference in distribution. This framework is a technique that optimizes the distribution difference between versions, transfer learning, and hyper-parameters of the classifier. We confirmed that the method of automatically selecting whether to perform transfer learning based on the distribution difference is effective through experiments. Moreover, we can see that using our optimization framework is effective in improving performance and, as a result, can reduce software inspection effort. This is expected to support practical quality assurance activities for new version projects in a cross-version project environment.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.1
• /
• pp.163-170
• /
• 2022

This paper introduces a method of extracting facial features due to unrefined inputs in real life and improving the problem of not guaranteeing the ideal performance and speed of the object recognition model through a storage algorithm through weight summation. Many facial recognition processes ensure accuracy in ideal situations, but the problem of not being able to cope with numerous biases that can occur in real life is drawing attention, which may soon lead to serious problems in the face recognition process closely related to security. This paper presents a method of quickly and accurately recognizing faces in real time by comparing feature points extracted as input with a small number of feature points that are not overfit to multiple biases, using that various variables such as picture composition eventually take an average form.

• Membrane Journal
• /
• v.20 no.2
• /
• pp.97-105
• /
• 2010

This study was to evaluate the performance of the Double stage Pore controllable fiber (DP) filter system as the pre-treatment of the RO membrane. The evaluation for the pre-treatment filter was performed through the indirect method, SDI (Silt Density Index) measurement of the filtrate. This study was done during Jan. 3 of 2009 to Dec. 3 of 2009 at OO Water Treatment Plant that was suppling industrial water to plants, and the raw water was contaminated lake water and it was fed to the system after clarification with coagulation. The average turbidity of the feed water and that of the filtrate was 0.79 NTU (0.28~4.01 NTU), and 0.16 NTU (0.04~0.50) respectively. And so the average turbidity removal efficiency was 77%. The filtrate flow rate and the backwash water flow rate was about 230 $m^3$/day and about 8.7 $m^3$/day respectively, and so the backwash rate was 3.8%. The data for some samples were obtained after a few days storage, and it caused the higher turbidity and SDI15 as the storage time was increased. But average SDI value of the filtrate was 3.6 (2.26~5.00) which was lower than minimum value required by the RO membrane manufacturer as the RO feed water to guarantee the life time of the RO membrane. So, the DP filter system was enough for the application as the pre-treatment of the RO membrane.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.846-851
• /
• 2005

In this paper, we propose Genetic Algorithms(GAs)-based Optimized Polynomial Neural Networks(PNN). The proposed algorithm is based on Group Method of Data Handling(GMDH) method and its structure is similar to feedforward Neural Networks. But the structure of PNN is not fixed like in conventional neural networks and can be generated in a dynamic manner. As each node of PNN structure, we use several types of high-order polynomial such as linear, quadratic and modified quadratic, and it is connected as various kinds of multi-variable inputs. The conventional PNN depends on the experience of a designer that select the number of input variables, input variable and polynomial type. Therefore it is very difficult to organize optimized network. The proposed algorithm leads to identify and select the number of input variables, input variable and polynomial type by using Genetic Algorithms(GAs). The aggregate performance index with weighting factor is proposed as well. The study is illustrated with tile NOx omission process data of gas turbine power plant for application to nonlinear process. In the sequel the proposed model shows not only superb predictability but also high accuracy in comparison to the existing intelligent models.

• Journal of the Korean Wood Science and Technology
• /
• v.27 no.1
• /
• pp.30-36
• /
• 1999

This paper describes the relationship between the bending strength properties of sloped finger-jointed woods and the acoustic emissions (AEs) generated during the test. Pinus densiflora pieces were cut in sloped-finger types and glued with four kinds of adhesives (polyvinyl acetate, polyvinyl-aeryl, oilic urethane and resorcinol-phenol resin). The results were as follows: The lower the bending strength(load) was, the earlier the generation time of AE event count got and the higher the increasing rate of AE event count became in the sloped finger-jointed specimens bonded with polyvinyl acetate and oilic urethane resin adhesives. Therefore, the slope from load-AE cumulative event count curve was very steep. The AE event count for resorcinol-phenol resin adhesive obtained even from low load level was abundant. The AE event count continuously increased as load increased and the event count was much more than one in the other conditions. The slope from load-AE cumulative event count curve was very gentle compared with other conditions. The patterns of AE event count and count were very similar. The relationship between the MOR and the AE parameter from load and AE cumulative event count in the early stage of the sloped finger-jointed specimens bonded with polyvinyl acetate, oilic urethane and resorcinol-phenol resin adhesives was much greater than that between the MOE and the MOR. Therefore, the AE signals obtained during bending test are useful for estimating the strength of sloped finger-jointed specimens.

• Journal of IKEEE
• /
• v.23 no.3
• /
• pp.1088-1091
• /
• 2019

In this paper, a real-time monitoring system for environmental safety by living area is proposed. The proposed system is designed to measure radiation, fine dust and basic living information (temperature) using fixed and mobile measuring equipment, and constitutes a web database that stores data received from the equipment. It also develops web programs for displaying received data on PCs and mobile phones. The results of testing the performance of the system by an authorized testing agency showed that the radiation measurement range was measured in the range of $10{\mu}Sv/h$ to 10mSv/h, which is comparable to the world's highest level, and that the accuracy was measured between ${\pm}6.7$ and ${\pm}8.7$ percent of the measurement uncertainty was measured and normal operation at or below the international standard of ${\pm}15$ percent. In addition, the temperature test was conducted on a section of $-20^{\circ}C$ to $50^{\circ}C$ and normal operation was confirmed in response to the temperature change. Stability of radiated electromagnetic waves was ensured by a suitable judgment. The product's testing in general and high and low temperature environments for about four months after the prototype was made confirmed to be more than five years of durability. The measurement range and accuracy of fine dust sensors are compared with those of companies that measure the air environment, and the performance level is similar through the air quality measurement register.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.5
• /
• pp.570-588
• /
• 2022

We scrutinized the acoustic outputs from the 70 shock wave generators of the 15 product models whose technical documents were available, among the 46 ballistic extracorporeal shock wave therapeutic devices of 11 domestic and 6 foreign manufacturers, approved by the Minster of Food & Drug Safety (Rep. Korea). We found that the acoustic Energy Flux Density (EFD), the most popular exposure parameter, was different by up to 563.64 times among shock wave generators at their minimum output settings and by up to 74.62 times at their maximum settings. In the same product model, the EFD was shown to vary depending on shock wave transmitters by up to 81.82 times at its minimum output setting and by up to 46.15 times at its maximum setting. The lowest EFD 0.013 mJ/mm² at the maximum output settings was much lower (2.1 %) than the maximum value 0.62 mJ/mm² at the minimum settings. The Large acoustic output differences (tens to hundreds of times)from the therapeutic devices approved for the same clinical indications imply that their therapeutic efficacy & safety may not be assured. The findings suggest the regulatory authority to revise her guideline to give clearer criteria for clinical approval and equality in performance, and recommend the authority to initiate a post-approval surveillance as well as a test in conformance between the data in technical documents and the real acoustic outputs clinically used.

• Composites Research
• /
• v.36 no.2
• /
• pp.132-139
• /
• 2023

A linerless composite propellant tank was designed and manufactured by using the carbon fiber-reinforced composite materials which have superior strength-to-weight ratio in order to reduce weight of the tank. In this research, we designed a sub-scale composite propellant tank with a diameter of 800 mm to withstand an MEOP of 1.7 MPa. We manufactured the boss of the tank by using the same composite materials to reduce the thermal expansion difference between the boss and the secondary-bonded composite layers of the barrel in the cryogenic environment. We used the collapsible mandrel to manufacture the tank without any liner. The mandrel was made from epoxy-based composite tooling prepregs to reduce weight of the mandrel. We manufactured the test tanks by laying up the carbon fiber fabric prepregs manually on the mandrel and then applying the autoclave cure process. We performed a proof test, a helium tightness test, a repeated pressurization test, and a burst test in room temperature. The test results demonstrate that the proposed design and manufacture process satisfies all strength requirements as well as an anti-leakage requirement.

• The KIPS Transactions:PartB
• /
• v.11B no.6
• /
• pp.749-758
• /
• 2004

In this paper, we propose a new method utilizing only raw corpus without additional human effort for disambiguation of target word selection in English-Korean machine translation. We use two data-driven techniques; one is the Latent Semantic Analysis(LSA) and the other the Probabilistic Latent Semantic Analysis(PLSA). These two techniques can represent complex semantic structures in given contexts like text passages. We construct linguistic semantic knowledge by using the two techniques and use the knowledge for target word selection in English-Korean machine translation. For target word selection, we utilize a grammatical relationship stored in a dictionary. We use k- nearest neighbor learning algorithm for the resolution of data sparseness Problem in target word selection and estimate the distance between instances based on these models. In experiments, we use TREC data of AP news for construction of latent semantic space and Wail Street Journal corpus for evaluation of target word selection. Through the Latent Semantic Analysis methods, the accuracy of target word selection has improved over 10% and PLSA has showed better accuracy than LSA method. finally we have showed the relatedness between the accuracy and two important factors ; one is dimensionality of latent space and k value of k-NT learning by using correlation calculation.

• Journal of the Korean Institute of Gas
• /
• v.11 no.1 s.34
• /
• pp.23-28
• /
• 2007

This paper presents an experimental investigation on the sealing defects and cracks of O-rings and a valve packing of a gas valve for a LPG cylinder. O-ring in which stops a gas leakage of a liquefied petroleum gas is very important for a LPG valve safety. Valve packing is to open and close a gas flow port for supplying and charging a LPG fuel. The sealing performance of two sealing units ism related to the leak safety and long lift of a gas valve. The investigated results show that most of O-rings was failed due to a circumferential crack in which is caused by partial press bonding failure near the partition zone and an excess compression rate. Some of the O-ring failure was originated by an extrusion of an excessive leak pressure of a LP gas. Thus, this paper strongly recommends a tight quality control and a safety guarantee system of O-rings and valve packing to guarantee a leak safety and to extend a service lift of a gas valve. At the end, a warranty policy of the sealing units should be adopted for increasing a product quality and safety of a gas valve.