• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.470-472
• /
• 2023

최근 위치 기반 장소 (POI) 추천 서비스가 많이 사용되면서, 사용자의 이전 방문지들에 대한 체크인 시퀀스를 기반으로 현재 (다음으로) 방문할 법한 POI 를 찾아 사용자에게 추천하는, next POI 추천 시스템에 관한 연구가 활발히 진행되고 있다. 하지만, 기존 연구들의 경우 next POI 추천을 위한 모델 학습 시, 사용자의 네거티브 POIs 에 관한 정교한 샘플링 없이 사용자 선호도를 추론해왔다. 본 연구에서는, 사전 학습된 별도의 사용자 선호도 추론 모델을 통해 사용자의 네거티브 POI로서 쉽게 분류되기 어려운 하드 네거티브 POIs 를 찾고, 이들을 위주로 수행되는 하드 네거티브 샘플링 방법을 새롭게 제안한다. 우리는 실 세계 데이터셋을 이용한 실험을 통해, 제안 방안이 기존 연구들에서 사용되어 온 랜덤 네거티브 샘플링 방법 대비 recall@5 기준, 최대 16.4%까지 추천 정확도를 향상시킬 수 있음을 확인하였다.

• Industry Promotion Research
• /
• v.8 no.1
• /
• pp.111-117
• /
• 2023

Sampling inspection methods for quality control have been proposed a lot in the industry. However, the sampling inspection method for reliability, which is a quality over time, has been relatively less presented, and there are not many literatures that are clearly summarized. Therefore, this paper focuses on the reliability conformity test to verify that the reliability evaluation scale value of the target is satisfied during the reliability test. To this end, first, we look at the conditions that both consumers and producers can satisfy in terms of the OC curve and find out what sampling methods satisfy the desired level of producer risk and consumer risk. Next, two methods of the reliability sampling methods such as attribute and variable reliability sampling methods are examined. Specifically, the attribute reliability sampling method is a form of sampling plan where n samples are tested for a certain period of T hours and the lot is accepted if the number of failures is less than or equal to a certain number c. On the other hand, the variable reliability sampling method is a form of sampling plan where the lot is accepted if the reliability evaluation scale such as MTBF satisfies a certain standard. Both sampling plans may also use inspection tables.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.10
• /
• pp.531-538
• /
• 2017

In the case of hydraulic pressure type fuse, we accept or reject certain product lots by considering the number of defective products in the operating pressure test. Generally, this procedure, known as 'The inspection by attributes', has been most commonly used in the field of quality assurance of products. However, the method of inspection by attributes suffers because it tests more samples than inspection by variables. Even though the quality of the products has remained stable in the process condition, the same number of samples is required for every lot, which wastes time and money. This paper suggests that the lot acceptance procedure is changed from inspection by attributes to inspection by variables. We can calculate the statistical tolerance percent of defectives and compare this to the Acceptable Quality Level (AQL) in order to save money and time. It is also easier to monitor and control the quality of products by using the process capability index and x-bar charts. In conclusion, the procedure delivers mutual benefit to both the customer and the producer by securing high quality products and reference data.

• Journal of Digital Contents Society
• /
• v.15 no.3
• /
• pp.347-355
• /
• 2014

The general octree structure is common for various applications including computer graphics, geometry information analysis and query. Unfortunately, the general octree approach causes duplicated sample data and discrepancy between sampling and representation positions when applied to sample continuous spatial information, for example, signed distance fields. To address these issues, some researchers introduced the dual octree. In this paper, the weakness of the dual octree approach will be illustrated by focusing on the fact that the dual octree cannot access some specific continuous zones asymptotically. This paper shows that the primal tree presented by Lefebvre and Hoppe can solve all the problems above. Also, this paper presents a three-dimensional primal tree traversal algorithm based the Morton codes which will help to parallelize the primal tree method.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.530-536
• /
• 2007

본 논문에서는 MFL(Magnetic flux leckage) 피그(PIG : Pipeline Inspection Gauge)에서 획득된 센서데이터의 분석을 위한 가시화 기법의 구현을 소개한다. MFL피그는 배관 내부에 삽입되어 배관의 결함이나 특징정보를 파악하기 위해 사용되는 장비로, 다양한 종류의 센서를 지니고 있으며, 각 센서에서 나온 값들은 피그에 탑재된 저장장치에 빠른 샘플링 속도로 저장된다. 분석가는 피그에 저장된 데이터를 가시화 도구를 사용하여, 피그에 샘플링 된 데이터를 통해 배관의 용접부위나 결함과 같은 특징정보를 찾아야 하고, 특징정보인 부분과 그렇지 않은 부분을 쉽게 구별 할 수 있어야 한다. 하지만 센서의 값에 따라 색상에 맵핑하여 보여주는 2차원 가시화 기법만으로는 효율적인 분석이 어렵다. 본 논문은 이를 극복하기 위하여, 센서 값에 맵핑되는 색상 스펙트럼 편집 및 기본 값 조정 기능을 제안한다.

• Journal of IKEEE
• /
• v.24 no.1
• /
• pp.178-185
• /
• 2020

This paper proposes precise time synchronization-based synchronized sampling structure applied HW/SW platform for LAN-based protection of future digital substations. The integrated software of the proposed platform includes IEC 61850 protocol, IEEE 1588 precision time protocol and synchronized sampling structure. The proposed platform expected to provide a basis of an application of future distributed sensing data-based protection and control methods by providing synchronized measurement among IEDs. The implementation of the proposed HW/SW platform technique was performed using TMDXIDK572 multi-core/multi-processor evaluation module and its time synchronization performance and synchronized sampling function were confirmed through the performance tests.

• Journal of Internet Computing and Services
• /
• v.17 no.2
• /
• pp.49-57
• /
• 2016

Ensemble classification involves combining multiple classifiers to obtain more accurate predictions than those obtained using individual models. Ensemble learning techniques are known to be very useful for improving prediction accuracy. Bagging is one of the most popular ensemble learning techniques. Bagging has been known to be successful in increasing the accuracy of prediction of the individual classifiers. Bagging draws bootstrap samples from the training sample, applies the classifier to each bootstrap sample, and then combines the predictions of these classifiers to get the final classification result. Bootstrap samples are simple random samples selected from the original training data, so not all bootstrap samples are equally informative, due to the randomness. In this study, we proposed a new method for improving the performance of the standard bagging ensemble by optimizing bootstrap samples. A genetic algorithm is used to optimize bootstrap samples of the ensemble for improving prediction accuracy of the ensemble model. The proposed model is applied to a bankruptcy prediction problem using a real dataset from Korean companies. The experimental results showed the effectiveness of the proposed model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.1
• /
• pp.218-224
• /
• 2013

Generally, SAW device, which uses Time Domain Sampling, requires high speed A/D converter because SAW device using TDS needs high sampling speed as much as its high data speed. However, the high price of A/D converter discourages makers from using it. On the other hand, SAW device, which uses Frequency Domain Sampling, does not required high speed A/D converter because SAW device using FDS does not need high sampling speed. It is very efficient in price comparison to its performance because high processing speed of SAW device using FDS can be implemented using low price Embedded Systems. The purpose of the paper is to solve the issues above by designing and realizing SAW device(FDS) using SAW sensor for TDS.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.1052-1055
• /
• 2020

본 논문에서는 3D 메시 정보, RGB-D 손 자세 및 2D/3D 손/세그먼트 마스크를 포함하여 인간의 손과 관련된 다양한 컴퓨터 비전 작업에 사용할 수 있는 새로운 다중 모달 합성 벤치마크를 제안 하였다. 생성된 데이터셋은 기존의 대규모 데이터셋인 BigHand2.2M 데이터셋과 변형 가능한 3D 손 메시(mesh) MANO 모델을 활용하여 다양한 손 포즈 변형을 다룬다. 첫째, 중복되는 손자세를 줄이기 위해 전략적으로 샘플링하는 방법을 이용하고 3D 메시 모델을 샘플링된 손에 피팅한다. 3D 메시의 모양 및 시점 파라미터를 탐색하여 인간 손 이미지의 자연스러운 가변성을 처리한다. 마지막으로, 다중 모달리티 데이터를 생성한다. 손 관절, 모양 및 관점의 데이터 공간을 기존 벤치마크의 데이터 공간과 비교한다. 이 과정을 통해 제안된 벤치마크가 이전 작업의 차이를 메우고 있음을 보여주고, 또한 네트워크 훈련 과정에서 제안된 데이터를 사용하여 RGB 기반 손 포즈 추정 실험을 하여 생성된 데이터가 양질의 질과 양을 가짐을 보여준다. 제안된 데이터가 RGB 기반 3D 손 포즈 추정 및 시맨틱 손 세그멘테이션과 같은 품질 좋은 큰 데이터셋이 부족하여 방해되었던 작업에 대한 발전을 가속화할 것으로 기대된다.

• Journal of Korea Spatial Information System Society
• /
• v.12 no.1
• /
• pp.18-27
• /
• 2010

In u-GIS environment, GeoSensor environment requires that dynamic data captured from various sensors and static information in terms of features in 2D or 3D are fused together. GeoSensors, the core of this environment, are distributed over a wide area sporadically, and are collected in any size constantly. As a result, storage space could be exceeded because of restricted memory in DSMS. To solve this kind of problems, a lot of related studies are being researched actively. There are typically 3 different methods - Random Load Shedding, Semantic Load Shedding, and Sampling. Random Load Shedding chooses and deletes data in random. Semantic Load Shedding prioritizes data, then deletes it first which has lower priority. Sampling uses statistical operation, computes sampling rate, and sheds load. However, they are not high accuracy because traditional ones do not consider spatial characteristics. In this paper 'Pre-Filtering based Post Load Shedding' are suggested to improve the accuracy of spatial query and to restrict load shedding in DSMS. This method, at first, limits unnecessarily increased loads in stream queue with 'Pre-Filtering'. And then, it processes 'Post-Load Shedding', considering data and spatial status to guarantee the accuracy of result. The suggested method effectively reduces the number of the performance of load shedding, and improves the accuracy of spatial query.