• 한국건설관리학회논문집
• /
• 제23권3호
• /
• pp.45-55
• /
• 2022

BIM의 도입에 따라 공간이 개별 객체로 인식되면서 객체화된 공간의 속성정보는 법규검토, 에너지 분석, 피난 경로 분석 등을 위한 기반 데이터로 사용 가능하기에 BIM의 활용성을 넓힐 수 있는 발판을 마련하였다. 그러나 BIM 모델 내 개별 공간 속성의 오기입이나 누락이 없는 시멘틱 무결성(semantic integrity)이 보장되어야 하는데, 다수의 참여자에 의한 수작업으로 진행되는 BIM 모델링 과정 특성 상 설계 오류가 빈번히 발생한다는 문제점이 존재한다. 이를 해결하기 위해 BIM 모델의 공간 정합성 검증을 위한 연구가 다수 진행되었으나, 적용 범위가 한정적이거나 분류 정확도가 낮은 한계점이 존재하였다. 본 연구에서는 공간의 기하정보 뿐 아니라 BIM 모델 내 공간과 부재 간 연결 관계를 Graph Convolutional Networks (GCN) 학습과정에 활용하여 향상된 성능의 공간 자동 분류모델을 구축하고자 하였다. 구축된 GCN 기반 모델의 성능을 공간의 기하정보만으로 학습된 기계학습 모델인 Multi-Layer Perceptron (MLP)과 비교하여 공간 분류 시 연결 관계 적용의 효용성을 검증하고자 하였다. 이를 통해 관계정보 활용 시 약 8% 내외 수준으로 공간 분류 성능이 향상되는 것으로 확인되었다.

• 인지과학
• /
• 제21권4호
• /
• pp.669-696
• /
• 2010

본 논문은 한 문서의 전체 의미는 각 부분의미의 합성이라는 관점에서 미리 반자동으로 구축된 감정어휘 평가사전을 기반으로 한 시스템을 제안한다. 인간의 의사 결정 과정과 유사한 방식으로 의사 결정 과정을 모델링하려는 노력으로써 본 ARSSA 시스템은 개별 리뷰의 의미값 연산과 자료 분류를 통해 감정 표현이 나타난 영화평 리뷰의 자동 등급화에 대한 연구를 수행한다. 이는 {'평점' : '리뷰'} 이항구조로 이루어진 현재의 평점 부여 형식에서 발생하는 두 변항의 불연속성 문제를 해결해보려는 목적을 가진다. 이는 어휘 의미 합성 과정에서 반영된 추상적 의미들의 합성 함수를 통해 실현될 수 있다. 시스템의 성능 실험에서 네이버 무비에서 확보한 1000개의 리뷰에 대한 10-fold 교차 검증 실험이 수행되었다. 이 실험은 기존에 부여된 평점과 비교하여 감정어휘 평가사전을 이용하였을 때 85%의 F1 Score를 보였다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제4권7호
• /
• pp.283-290
• /
• 2015

본 논문에서는 WiseQA 시스템에서 정답유형을 인식하기 위한 하이브리드 방법을 제안한다. 정답유형은 어휘정답유형과 의미정답유형으로 구분된다. 본 논문은 어휘정답유형 인식을 위해서 질문초점에 기반한 규칙모델과 순차적 레이블링에 기반한 기계학습모델을 제안한다. 의미정답유형 인식을 위해 다중클래스 분류에 기반한 기계학습모델과 어휘정답유형을 이용한 필터링 규칙을 소개한다. 어휘정답유형 인식성능은 F1-score 82.47%이고, 의미정답유형 인식성능은 정확률 77.13%이다. 어휘정답유형 인식성능은 IBM 왓슨과 비교하여, 정확률은 1.0% 저조하고, 재현율은 7.4% 높다.

• 한국안전학회지
• /
• 제34권6호
• /
• pp.50-57
• /
• 2019

In this study, various types of deep learning models that have been proposed recently are classified according to data input / output types and analyzed to find the deep learning model suitable for constructing a crack detection model. First the deep learning models are classified into image classification model, object segmentation model, object detection model, and instance segmentation model. ResNet-101, DeepLab V2, Faster R-CNN, and Mask R-CNN were selected as representative deep learning model of each type. For the comparison, ResNet-101 was implemented for all the types of deep learning model as a backbone network which serves as a main feature extractor. The four types of deep learning models were trained with 500 crack images taken from real concrete structures and collected from the Internet. The four types of deep learning models showed high accuracy above 94% during the training. Comparative evaluation was conducted using 40 images taken from real concrete structures. The performance of each type of deep learning model was measured using precision and recall. In the experimental result, Mask R-CNN, an instance segmentation deep learning model showed the highest precision and recall on crack detection. Qualitative analysis also shows that Mask R-CNN could detect crack shapes most similarly to the real crack shapes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권10호
• /
• pp.3230-3255
• /
• 2022

Causality mining in NLP is a significant area of interest, which benefits in many daily life applications, including decision making, business risk management, question answering, future event prediction, scenario generation, and information retrieval. Mining those causalities was a challenging and open problem for the prior non-statistical and statistical techniques using web sources that required hand-crafted linguistics patterns for feature engineering, which were subject to domain knowledge and required much human effort. Those studies overlooked implicit, ambiguous, and heterogeneous causality and focused on explicit causality mining. In contrast to statistical and non-statistical approaches, we present Bidirectional Encoder Representations from Transformers (BERT) integrated with Multi-level Feature Networks (MFN) for causality recognition, called BERT+MFN for causality recognition in noisy and informal web datasets without human-designed features. In our model, MFN consists of a three-column knowledge-oriented network (TC-KN), bi-LSTM, and Relation Network (RN) that mine causality information at the segment level. BERT captures semantic features at the word level. We perform experiments on Alternative Lexicalization (AltLexes) datasets. The experimental outcomes show that our model outperforms baseline causality and text mining techniques.

• 한국BIM학회 논문집
• /
• 제13권4호
• /
• pp.96-105
• /
• 2023

This study focuses on developing a building object recognition technology for efficient use in the remodeling of buildings constructed without drawings. In the era of the 4th industrial revolution, smart technologies are being developed. This research contributes to the architectural field by introducing a deep learning-based method for automatic object classification and recognition, utilizing point cloud data. We use a TD3D network with voxels, optimizing its performance through adjustments in voxel size and number of blocks. This technology enables the classification of building objects such as walls, floors, and roofs from 3D scanning data, labeling them in polygonal forms to minimize boundary ambiguities. However, challenges in object boundary classifications were observed. The model facilitates the automatic classification of non-building objects, thereby reducing manual effort in data matching processes. It also distinguishes between elements to be demolished or retained during remodeling. The study minimized data set loss space by labeling using the extremities of the x, y, and z coordinates. The research aims to enhance the efficiency of building object classification and improve the quality of architectural plans by reducing manpower and time during remodeling. The study aligns with its goal of developing an efficient classification technology. Future work can extend to creating classified objects using parametric tools with polygon-labeled datasets, offering meaningful numerical analysis for remodeling processes. Continued research in this direction is anticipated to significantly advance the efficiency of building remodeling techniques.

• 한국데이타베이스학회:학술대회논문집
• /
• 한국데이타베이스학회 2008년도 연합학회학술대회
• /
• pp.413-419
• /
• 2008

상품 정보 시스템은 상품의 분류정보, 관계정보, 속성정보 등을 가지고 있는 시스템으로, 현재의 상품 정보 시스템은 그 정보를 사용하는 서비스에 종속된 형태의 관계형 데이터베이스로 만들어져 있다. 본 논문에서는 시맨틱 웹 환경에서 상호 운영성을 확보하여 다양한 서비스에 적용할 수 있는 트리플 형태의 상품 정보 모델을 제시하며, 그 모델을 이용한 웹 서비스 방법을 제시한다.

• 한국정보처리학회논문지
• /
• 제7권7호
• /
• pp.2108-2115
• /
• 2000

The conventional Boolean retrieval systems based on vector spae model can provide the results of retrieval fast, they can't reflect exactly user's retrieval purpose including semantic information. Consequently, the results of retrieval process are very different from those users expected. This fact forces users to waste much time for finding expected documents among retrieved documents. In his paper, we designed a bayesian SOM(Self-Organizing feature Maps) in combination with bayesian statistical method and Kohonen network as a kind of unsupervised learning, then perform classifying documents depending on the semantic similarity to user query in real time. If it is difficult to observe statistical characteristics as there are less than 30 documents for clustering, the number of documents must be increased to at least 50. Also, to give high rank to the documents which is most similar to user query semantically among generalized classifications for generalized clusters, we find the similarity by means of Kohonen centroid of each document classification and adjust the secondary rank depending on the similarity.

• 가정과삶의질연구
• /
• 제11권1호
• /
• pp.159-175
• /
• 1993

The purpose of this study was to find out the psychological distance through semantic app-roach perceived by adolescent child in the subtypes of 'Extrem Family' dysfunctional families by classification of Olson and his associates ' Circrumplex Model. The subjects of this research were 1072 abolescents living in Seoul. Korea The survey methods were questionnaires including FACES II and The Psychological Distance Scale. Data were analyzed by means of the statistics of frequency percentage arithematic mean standard devia-tion crosstabs and one way-anova. The major findings are as follows: 1) The levels of family cohesion family adaptibility and the psychologival distances with father mother and siblings perceived by adolescent were high. 2) The most of subject's families belonged to 'Balanced Family' in the types of family system ' Extreme Family' type showed the lowest frequency and the main subtypes of it that had the highest frequency were 'Enmeshed Chaotic Family' ' Disengaged Rigid Family' 3) While adolescents of 'Enmeshed Chaotic Family' perceived most closely with other family members. those of 'Disengated Rigid Family' most distantly totally and in evaluation potency and activity three subfactors in psychological distance. 4) There were differences of unit points in subfactors of psychological distances with other family members perceived by adolescents according to the types of family system. While the points of 'Enmeshed Chaotic Family' were the highest those of 'Disengaged Rigid Family' were the lowest. 5) While 'Enmeshed Chaotic Family' were located most closely 'Disengaged Rigid Family' were located most distantly in the mutual distances and direct distances among family concepts on semantic space.

• 대한인간공학회지
• /
• 제23권4호
• /
• pp.9-21
• /
• 2004

As menu structure of household appliance is complicated, user's cognitive workload frequently occurs errors. In existing studies, errors didn't present that interpretation for cognitive factors and alternatives, but are only considered as statistical frequency. Therefore, error classification and analysis in tasks is inevitable in usability evaluation. This study classified human error throughout information process model and navigation behavior. Human error is defined as incorrect decision and behavior reducing performance. And navigation is defined as unrelated behavior with target item searching. We searched and analyzed human errors and its causes as a case study, using mobile phone which could control appliances in near future. In this study, semantic problems in menu structure were elicited by SAT. Scenarios were constructed by those. Error analysis tests were performed twice to search and analyze errors. In 1st prototype test, we searched errors occurred in process of each scenario. Menu structure was revised to be based on results of error analysis. Henceforth, 2nd Prototype test was performed to compare with 1st. Error analysis method could detect not only mistakes, problems occurred by semantic structure, but also slips by physical structure. These results can be applied to analyze cognitive causes of human errors and to solve their problems in menu structure of electronic products.