• 한국군사과학기술학회지
• /
• 제14권4호
• /
• pp.719-725
• /
• 2011

To apply TMAH-based Py-MS to a field biological detection system for real-time classification of cell-type, reproducible patterns of the TMAH-based Py-MS spectra was known as a critical factor for classification but was seriously disturbed by quantity of cells injected into pyro-tube. This factor is an exterior variable that could not be complemented by improving the performance of the TMAH-based Py-MS instrument. One of idea to solve the knotty problem has been flashed from "Top-down proteomics for identification of intact microoganisms". That is, biomarker peaks are selected from complicate Py-MS spectra for intact microoganisms by tracing out their origins, based on Py-MS spectra for the featured components of different cell-types, in Top-down approach. This idea has been tested in classification of different Gram-type microoganisms. Through the analyses of spectra for the featured components - peptidoglycan and lipoteichoic acid for Gram-positive cells and lipopolysaccharide and lipid A for Gram-negative cells - with comparing to the spectra the corresponding Gram-type cells in the Top-down approach, biomarker peaks were selected to carry out PCA(Principal Component Analysis) in order to see classification of different Gram-types, resulting in significant improvement of their classification. Furthermore, weighting biomarker peaks on intact cell's spectra, based on the data for the featured components of the Gram-types, contributed to elevate classification performance.

• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.803-816
• /
• 2023

Explainable artificial intelligence is a method that explains how a complex model (e.g., a deep neural network) yields its output from a given input. Recently, graph-type data have been widely used in various fields, and diverse graph neural networks (GNNs) have been developed for graph-type data. However, methods to explain the behavior of GNNs have not been studied much, and only a limited understanding of GNNs is currently available. Therefore, in this paper, we propose an explanation method for node classification using graph convolutional networks (GCNs), which is a representative type of GNN. The proposed method finds out which features of each node have the greatest influence on the classification of that node using GCN. The proposed method identifies influential features by backtracking the layers of the GCN from the output layer to the input layer using the gradients. The experimental results on both synthetic and real datasets demonstrate that the proposed explanation method accurately identifies the features of each node that have the greatest influence on its classification.

• 한국산학기술학회논문지
• /
• 제5권6호
• /
• pp.539-545
• /
• 2004

자연어 질의를 입력하고 문서로부터 질의에 대한 정답을 추출하여 제공하는 질의응답 시스템에서는 사용자의 질의 의도를 파악하여 질의 유형을 분류하는 과정이 매우 중요하다. 본 논문에서는 질의 유형을 분류하기 위해 복잡한 분류 규칙이나 대용량의 사전 정보를 이용하지 않고 질의의 의도를 나타내는 어휘들을 추출하고 인접 명사들의 의미 정보를 이용하여 질의 및 정답 유형을 결정할 수 있는 방법을 제안한다. 또 동의어 정보와 접미사 정보를 이용하고, 의문사가 생략된 경우 어휘 의미 정보를 이용하여 질의 유형 분류기의 성능을 향상시킬 수 있음을 보인다.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2003년도 봄 학술발표회 논문집
• /
• pp.265-272
• /
• 2003

In this study, the rock mass classification results from the face mapping and the resistivity inversion data are compared and analyzed for the reliability investigation of the determination of the rock support type based on the surface electrical survey. To get the quantitative correlation, rock engineering indices such as RCR(rock condition rating), N(Rock mass number), Q-system based on RMR(rock mass rating) are calculated. Kriging method as a post processing technique for global optimization is used to improve its resolution. The result of correlation analysis shows that the geological condition estimated from 2D electrical resistivity survey is coincident globally with the trend of rock type except for a few local areas. The correlation between the results of 3D electrical resistivity survey and the rock mass classification turns out to be very high. It can be concluded that 3D electrical resistivity survey is powerful to set up the reliable rock support type.

• 대한임베디드공학회논문지
• /
• 제10권3호
• /
• pp.179-188
• /
• 2015

We propose obstacle classification method based on 2D LIDAR(Light Detecting and Ranging) database. The existing obstacle classification method based on 2D LIDAR, has an advantage in terms of accuracy and shorter calculation time. However, it was difficult to classifier the type of obstacle and therefore accurate path planning was not possible. In order to overcome this problem, a method of classifying obstacle type based on width data of obstacle was proposed. However, width data was not sufficient to improve accuracy. In this paper, database was established by width, intensity, variance of range, variance of intensity data. The first classification was processed by the width data, and the second classification was processed by the intensity data, and the third classification was processed by the variance of range, intensity data. The classification was processed by comparing to database, and the result of obstacle classification was determined by finding the one with highest similarity values. An experiment using an actual autonomous vehicle under real environment shows that calculation time declined in comparison to 3D LIDAR and it was possible to classify obstacle using single 2D LIDAR.

• 대한원격탐사학회지
• /
• 제34권1호
• /
• pp.141-150
• /
• 2018

Crop type classification is essential for supporting agricultural decisions and resource monitoring. Remote sensing techniques, especially using hyperspectral imagery, have been effective in agricultural applications. Hyperspectral imagery acquires contiguous and narrow spectral bands in a wide range. However, large dimensionality results in unreliable estimates of classifiers and high computational burdens. Therefore, reducing the dimensionality of hyperspectral imagery is necessary. In this study, the Random Forest (RF) classifier was utilized for dimensionality reduction as well as classification purpose. RF is an ensemble-learning algorithm created based on the Classification and Regression Tree (CART), which has gained attention due to its high classification accuracy and fast processing speed. The RF performance for crop classification with airborne hyperspectral imagery was assessed. The study area was the cultivated area in Chogye-myeon, Habcheon-gun, Gyeongsangnam-do, South Korea, where the main crops are garlic, onion, and wheat. Parameter optimization was conducted to maximize the classification accuracy. Then, the dimensionality reduction was conducted based on RF variable importance. The result shows that using the selected bands presents an excellent classification accuracy without using whole datasets. Moreover, a majority of selected bands are concentrated on visible (VIS) region, especially region related to chlorophyll content. Therefore, it can be inferred that the phenological status after the mature stage influences red-edge spectral reflectance.

• 한국컴퓨터정보학회논문지
• /
• 제25권2호
• /
• pp.213-219
• /
• 2020

본 연구는 스마트폰 과의존을 진단하고 예측하기 위하여 할 수 있는 분류분석 방법과 스마트폰 과의존 분류율에 영향을 미치는 중요변수를 규명하고자 시도되었다. 이를 위해 인공지능의 방법인 기계학습 분석 기법 중 의사결정트리, 랜덤포레스트, 서포트벡터머신의 분류율을 비교하였다. 자료는 한국정보화진흥원에서 제공한 '2018년 스마트폰 과의존 실태조사'에 응답한 25,465명의 데이터였고, R 통계패키지(ver. 3.6.2)를 사용하여 분석하였다. 분석한 결과, 3가지 분류분석 기법은 정분류율이 유사하게 나타났으며, 모델에 대한 과적합 문제가 발생되지 않았다. 3가지 분류분석 방법 중 서포트벡터머신의 분류율이 가장 높게 나타났고, 다음으로 의사결정트리 기법, 랜덤포레스트 기법 순이었다. 스마트폰 이용 유형 중 분류율에 영향을 미치는 상위 3개 변수는 생활서비스형, 정보검색형, 여가추구형이었다.

• Journal of Periodontal and Implant Science
• /
• 제51권2호
• /
• pp.100-113
• /
• 2021

Purpose: Previous studies have solely focused on fresh extraction sockets, whereas in clinical settings, alveolar sockets are commonly associated with chronic inflammation. Because the extent of tissue destruction varies depending on the origin and the severity of inflammation, infected alveolar sockets may display various configurations of their remaining soft and hard tissues following tooth extraction. The aim of this study was to classify infected alveolar sockets and to provide the appropriate treatment approaches. Methods: A proposed classification of extraction sockets with chronic inflammation was developed based upon the morphology of the bone defect and soft tissue at the time of tooth extraction. The prevalence of each type of the suggested classification was determined retrospectively in a cohort of patients who underwent, between 2011 and 2015, immediate bone grafting procedures (ridge preservation/augmentation) after tooth extractions at Seoul National University Dental Hospital. Results: The extraction sockets were classified into 5 types: type I, type II, type III, type IV (A & B), and type V. In this system, the severity of bone and soft tissue breakdown increases from type I to type V, while the reconstruction potential and treatment predictability decrease according to the same sequence of socket types. The retrospective screening of the included extraction sites revealed that most of the sockets assigned to ridge preservation displayed features of type IV (86.87%). Conclusions: The present article classified different types of commonly observed infected sockets based on diverse levels of ridge destruction. Type IV sockets, featuring an advanced breakdown of alveolar bone, appear to be more frequent than the other socket types.

• Journal of Chest Surgery
• /
• 제38권1호
• /
• pp.44-49
• /
• 2005

흥선종양은 비교적 흔한 종격종 종양이나 이제까지 병리학적 분류가 통일된 것이 없었으며 또한 치료 및 예후와의 연관성이 잘 확립되어 있지 않았다. 최근에서야 WHO 분류가 발표되었고 이에 따른 치료 계획과 치료에 따른 예후와의 상관관계가 보고되기 시작했다. 본 연구는 WHO 분류와 Masaoka병기 그리고 임상양상 간의 상관관계를 조사하였다. 대상 및 방법: 대상환자는 서울아산병원 흉부외과에서 1993년 1월부터 2003년 6월까지 완전절제술을 시행 받았던 흥선종양 환자 98명으로 하였다. WHO 분류의 조사를 위하여 병리조직 slide를 다시 검토하였으며 수술 후 Masaoka병기와의 관련성, 술 후 추가적인 치료와 예후에 대한 관계 및 재발여부에 관하여 의무기록 조사를 통하여 후향적으로 조사하였다. 결과: 98예의 대상 환자 중 남녀 비는 48 : 50이었으며 수술 연령은 평균 $49.6{\pm}13.9$세였다. WHO 분류에 따르면 type A 6명, AB 14명, B1 18명, B2 23명, B3 18명, C 9명이었다. Masaoka 병기와 WHO 분류와의 관계를 보면 Masaoka 병기 I 53명 $(54{\%})$ 중에서 WHO type A 4명, AB 7명, B1 22명, B2 17명, B3 3명이었으며 Masaoka 병기 II 28명$(28.5{\%})$ 중에서는 WHO type A 2명, AB 7명, B1 4명, B2 2명, B3 8명, C 5명이었고 Masaoka병기 III 15명$(15.3{\%})$ 중에서는 WHO type B1 L명, B2 3명, B3 7명, C 4명이었으며 Masaoka병기 IV 2명$(2{\%})$ 중에서는 WHO type B1 1명, B2 1명이었다. 평균 추적 기간은 $28{\pm}6.8$개월이었다. 사망 환자는 3명으로 type B2에서 2명(Masaoka 병기 III, IV), 그리고 type C에서 1명(Masaoka병기 II)이었다 재발 후 생존해 있는 환자는 총 6명이었으며 이 중 type B2에서 2명(Masaoka 병기 III), type B3에서 2명(Masaoka 병기 I, III) type C에서 2명(Masaoka 병기 II)이었다. Kaplan-Meier방법으로 통계 처리한 결과 WHO분류상 type B2에서 5년 생존율은 $90{\%}$ 였으며 type C에서 5년 생존율은 $87.5{\%}$였다. 재발률을 보면 class B2에서 5년 무병 생존율 $80.7{\%}$, B3에서 $81.6{\%}$, C에서 $50{\%}$였다. Log-Rank 방법에서 보면 WHO분류와 생존율, 재발률 사이에 통계학적으로 상관관계가 있는 것으로 나타났다(p<0.05). WHO 분류와 Masaoka분류의 상관 관계를 보면 Spearman correction method출 이용한 통계에서 상관관계 곡선이 slope=0.401 (p=0.023)으로 밀접한 관계가 있다고 하겠다. 결론: WHO분류의 type C의 경우 수술 후 재발률과 사망률이 높으므로 수술 후보다 적극적인 치료와 추적관찰이 필요할 것으로 생각된다. WHO분류와 Masaoka 병기간에는 상호 밀접한 관계가 있는 것으로 생각되며 WHO분류 및 Masaoka병기 모두 흥선종의 예후의 예측 인자가 될 수 있을 것으로 생각된다.

• 대한수학회지
• /
• 제35권2호
• /
• pp.315-330
• /
• 1998

Chen and Piccinni [7] have classified all compact surfaces in a Euclidean space $R^{2＋p}$ with 1-type generalized Gauss map. Being motivated by this result, the purpose of this paper is to consider the Lorentz version of the classification theorem and to obtain a complete classification of space-like surfaces in indefinite Euclidean space $R_{p}$ $^{2＋p}$ with 1-type generalized Gauss map.p.