• Geomechanics and Engineering
• /
• 제25권1호
• /
• pp.1-16
• /
• 2021

Machine learning models have been widely used for landslide susceptibility assessment (LSA) in recent years. The large number of inputs or conditioning factors for these models, however, can reduce the computation efficiency and increase the difficulty in collecting data. Feature selection is a good tool to address this problem by selecting the most important features among all factors to reduce the size of the input variables. However, two important questions need to be solved: (1) how do feature selection methods affect the performance of machine learning models? and (2) which feature selection method is the most suitable for a given machine learning model? This paper aims to address these two questions by comparing the predictive performance of 13 feature selection-based machine learning (FS-ML) models and 5 ordinary machine learning models on LSA. First, five commonly used machine learning models (i.e., logistic regression, support vector machine, artificial neural network, Gaussian process and random forest) and six typical feature selection methods in the literature are adopted to constitute the proposed models. Then, fifteen conditioning factors are chosen as input variables and 1,017 landslides are used as recorded data. Next, feature selection methods are used to obtain the importance of the conditioning factors to create feature subsets, based on which 13 FS-ML models are constructed. For each of the machine learning models, a best optimized FS-ML model is selected according to the area under curve value. Finally, five optimal FS-ML models are obtained and applied to the LSA of the studied area. The predictive abilities of the FS-ML models on LSA are verified and compared through the receive operating characteristic curve and statistical indicators such as sensitivity, specificity and accuracy. The results showed that different feature selection methods have different effects on the performance of LSA machine learning models. FS-ML models generally outperform the ordinary machine learning models. The best FS-ML model is the recursive feature elimination (RFE) optimized RF, and RFE is an optimal method for feature selection.

• 한국CDE학회논문집
• /
• 제16권3호
• /
• pp.178-186
• /
• 2011

Static CAD models are the CAD models that do not have feature information and modeling history. These static models are generated by translating CAD models in a specific CAD system into neutral formats such as STEP and IGES. When a CAD model is translated into a neutral format, its precious feature information such as feature parameters and modeling history is lost. Once the feature information is lost, the advantage of feature based modeling is not valid any longer, and modification for the model is purely dependent on geometric and topological manipulations. However, the capabilities of the existing methods to modify static CAD models are limited, Direct modification methods such as tweaking can only handle the modifications that do not involve topological changes. There was also an approach to modify static CAD model by using volume decomposition. However, this approach was also limited to modifications of protrusion features. To address this problem, we extend the volume decomposition approach to handle not only protrusion features but also depression features in a static CAD model. This method first generates the model that contains the volume of depression feature using the bounding box of a static CAD model. The difference between the model and the bounding box is selectively decomposed into so called the feature volume and the base volume. A modification of depression feature is achieved by manipulating the feature volume of the static CAD model.

• Journal of information and communication convergence engineering
• /
• 제21권1호
• /
• pp.82-89
• /
• 2023

Various machine-learning models may yield high predictive power for massive time series for time series prediction. However, these models are prone to instability in terms of computational cost because of the high dimensionality of the feature space and nonoptimized hyperparameter settings. Considering the potential risk that model training with a high-dimensional feature set can be time-consuming, we evaluate a feature-importance-based feature selection method to derive a tradeoff between predictive power and computational cost for time series prediction. We used two machine learning techniques for performance evaluation to generate prediction models from a retail sales dataset. First, we ranked the features using impurity- and Local Interpretable Model-agnostic Explanations (LIME) -based feature importance measures in the prediction models. Then, the recursive feature elimination method was applied to eliminate unimportant features sequentially. Consequently, we obtained a subset of features that could lead to reduced model training time while preserving acceptable model performance.

• 한국측량학회지
• /
• 제39권5호
• /
• pp.297-311
• /
• 2021

Around the world, there is an increasing interest in Digital Twin cities. Although geospatial data is critical for building a digital twin city, currently-established spatial data cannot be used directly for its implementation. Integration of geospatial data is vital in order to construct and simulate the virtual space. Existing studies for data integration have focused on data transformation. The conversion method is fundamental and convenient, but the information loss during this process remains a limitation. With this, standardization of the data model is an approach to solve the integration problem while hurdling conversion limitations. However, the standardization within indoor space data models is still insufficient compared to 3D building and city models. Therefore, in this study, we present a comparative analysis of data models commonly used in indoor space modeling as a basis for establishing a generic indoor space feature model. By comparing five models of IFC (Industry Foundation Classes), CityGML (City Geographic Markup Language), AIIM (ArcGIS Indoors Information Model), IMDF (Indoor Mapping Data Format), and OmniClass, we identify essential elements for modeling indoor space and the feature classes commonly included in the models. The proposed generic model can serve as a basis for developing further indoor feature models through specifying minimum required structure and feature classes.

• 인터넷정보학회논문지
• /
• 제10권4호
• /
• pp.105-113
• /
• 2009

특성 모델(Feature Model)은 소프트웨어 제품 라인 개발 시 도메인 공학 단계에서 제품들 사이의 공통점과 차이점을 모델링하는데 널리 사용된다. 특성 모델의 오류 또는 불일치성에 대한 발견 및 수정은 성공적인 소프트웨어 제품 라인 공학을 위해서 필수적이다. 특성 모델의 검증을 효과적으로 수행하기 위해서는 자동화된 도구의 도움이 필요하다. 본 논문에서는 JESS 규칙 기반 시스템을 이용하여 특성 모델의 유효성을 검증하는 기법을 기술하고 이를 이용한 특성 모델 검증 도구를 제안한다. 본 논문의 도구는 특성 모델링 작업 시 실시간으로 특성 모델을 검증하여 오류의 존재 여부와 오류의 원인에 대한 설명을 제공함으로써 오류 없는 특성 모델을 생성할 수 있도록 해 준다. 특성 모델 검증 기법에 규칙 기반 시스템을 이용한 경우는 본 논문이 최초의 시도로 사료된다.

• 한국CDE학회논문집
• /
• 제10권6호
• /
• pp.444-454
• /
• 2005

Recently, the requirements of multi-resolution models of a solid model, which represent an object at multiple levels of feature detail, are increasing for engineering tasks such as analysis, network-based collaborative design, and virtual prototyping and manufacturing. The research on this area has focused on several topics: topological frameworks for representing multi-resolution solid models, criteria for the level of detail (LOD), and generation of valid models after rearrangement of features. As a solution to the feature rearrangement problem, the new concept of the effective zone of a feature is introduced in the former part of the paper. In this paper, we propose a feature-based non-manifold modeling system to provide multi-resolution models of a feature-based solid or non-manifold model on the basis of the effective feature zones. To facilitate the implementation, we introduce the class of the multi-resolution feature whose attributes contain all necessary information to build a multi-resolution solid model and extract LOD models from it. In addition, two methods are introduced to accelerate the extraction of LOD models from the multi-resolution modeling database: the one is using an NMT model, known as a merged set, to represent multi-resolution models, and the other is storing differences between adjacent LOD models to accelerate the transition to the other LOD. We also suggest the volume of the feature, regardless of feature type, as a criterion for the LOD. This criterion can be used in a wide range of applications, since there is no distinction between additive and subtractive features unlike the previous method.

• 인터넷정보학회논문지
• /
• 제11권4호
• /
• pp.107-117
• /
• 2010

소프트웨어 제품들 사이의 공통된 개념과 서로 다른 개념들을 표현한 특성 모델과, 특정 제품에 포함될 특성들을 선택한 결과인 특성 구성은 소프트웨어 프러덕트 라인 개발 방법론에서 핵심 요소이다. 이들에 대한 정형적 시맨틱과 논리적 추론에 대한 연구가 진행 중이지만 시맨틱 웹 기술을 이용한 특성 모델 온톨로지 구축과 특성 구성 검증에 대한 연구는 아직 부족한 상황이다. 본 논문에서는 온톨로지와 시맨틱 웹 기술을 이용하여 특성 모델의 정형적 시맨틱을 정의하고 특성 구성을 검증하는 기법을 제안한다. 특성 모델과 특성 구성에 포함된 지식을 시맨틱 웹 표준 언어인 OWL(Web Ontology Language)로 표현하고 특성 구성을 검증하기 위한 규칙은 시맨틱 웹 규칙 언어인 SWRL(Semantic Web Rule Language)로 정의한다. 본 논문의 기법은, 특성 모델의 정형적 시맨틱을 제공하며 특성 구성 검증을 자동화할 뿐 만 아니라 SQWRL과 같은 다양한 시맨틱 웹 기술 적용을 가능하게 한다.

• 스마트미디어저널
• /
• 제10권3호
• /
• pp.9-22
• /
• 2021

Basically, machine learning models use input data to produce results. Sometimes, the input data is too complicated for the models to learn useful patterns. Therefore, feature engineering is a crucial data preprocessing step for constructing a proper feature set to improve the performance of such models. One of the most efficient methods for automating feature engineering is the autoencoder, which transforms the data from its original space into a latent space. However certain factors, including the datasets, the machine learning models, and the number of dimensions of the latent space (denoted by k), should be carefully considered when using the autoencoder. In this study, we design a framework to compare two data preprocessing approaches: with and without autoencoder and to observe the impact of these factors on autoencoder. We then conduct experiments using autoencoders with classifiers on popular datasets. The empirical results provide a perspective regarding the best suited autoencoder for these factors.

• 한국IT서비스학회지
• /
• 제10권3호
• /
• pp.189-201
• /
• 2011

Feature models are widely used to model commonalities and variabilities among products during software product line development. Feature configurations are generated by selecting the features to be included in individual products. Automated tools to identify errors or inconsistencies in the feature models and configurations are essential to successful software product line engineering. This paper proposes a verification technique and tool based on semantic web technologies such as OWL, SWRL and Protege API. This approach checks the feature model and configuration based on predefined rules and provides information on existence of errors as well as the kinds of those errors. This approach is extensible due to ease of rule modification and may be easily applied to other environments because semantic web technologies can be easily integrated with other programming environments. This paper demonstrates how various semantic web-related technologies can support automatic verification of one kind of software development artifact, the feature model.

• 한국CDE학회논문집
• /
• 제13권2호
• /
• pp.131-138
• /
• 2008

With the advancements of the Internet and CAD data translation techniques, more CAD models are transferred from a CAD system to another through the network and interoperability is getting a common word in the CAD industry. However, when a CAD model is translated for an incompatible system into a neutral format such as STEP or IGES, its precious feature information is lost. When this feature information is lost, the advantage of feature based modeling is not valid any longer, and modification for the model is purely dependent on geometric and topological manipulations. However, the capabilities of the existing methods to modify these feature-independent models are limited as the modification involves a topological change in the model. To address this issue, we present a volumetric method to modify the solid models in neutral format. First, this method selectively decomposes the solid model to separate the portion of interest called feature volume. Next, the designer modifies the feature volume without concerning a topological change. Finally, the feature volume is united with the original solid model to complete the modification process. The results of test cases are presented to attest the usefulness of the proposed method.