• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.450-452
• /
• 2020

기술의 발전과 함께 사용자에게 가까이 자리 잡은 소셜 네트워크 서비스는 이미지, 동영상, 텍스트 등 활용 가능한 데이터의 수를 폭발적으로 증가시켰다. 작성자의 감정을 포함하고 있는 텍스트 데이터는 시장 조사, 주가 예측 등 다양한 분야에서 이용할 수 있으며, 이로 인해 긍부정의 이진 분류가 아닌 다중 감정 분석의 필요성 또한 높아지고 있다. 본 논문에서는 딥러닝 기반 감정 분류에 심리학 이론의 기반 감정 모델을 활용한 결합 모델과 단일 모델을 비교한다. 학습을 위해 AI Hub에서 제공하는 데이터와 노래 가사 데이터를 복합적으로 사용하였으며, 결과에서는 대부분의 경우에 결합 모델이 높은 결과를 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.495-497
• /
• 2022

COVID-19 상황이 지속됨에 따라 플라스틱 쓰레기 배출량은 해마다 기하급수적으로 증가하고 있는 반면 플라스틱 폐기물의 재활용률은 현저히 낮은 편에 속한다. 이러한 문제점들을 해결하기 위해 국가적으로 여러 플라스틱 폐기물 중 순환 가능한 PET를 분리하여 수거하고자 하는 노력을 하고 있다. 하지만, 현재 대량의 플라스틱 폐기물은 수거되는 시점부터 여러 폐기물과 혼합된 형태로 재활용 센터에 수거되어 추가 분류하는 인적자원이 요구되는 문제점이 존재한다. 따라서 본 논문에서는 이러한 한계점들을 해결하기 위해 AI 기술 중 하나인 Multi-Object Detection의 YOLOv7 모델을 적용하여 실시간으로 PET에 부착된 객체들을 탐지함으로써 순환 가능한 PET만을 분류하는 YOLOv7 기반 순환 가능한 PET 분류시스템을 설계 및 구현한다.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.13 no.5
• /
• pp.894-902
• /
• 2010

This paper presents a vision-based robust off-road terrain classification method against environmental variation. As a supervised classification algorithm, we applied a neural network classifier using wavelet features extracted from wavelet transform of an image. In order to get over an effect of overall image feature variation, we adopted environment sensors and gathered the training parameters database according to environmental conditions. The robust terrain classification algorithm against environmental variation was implemented by choosing an optimal parameter using environmental information. The proposed algorithm was embedded on a processor board under the VxWorks real-time operating system. The processor board is containing four 1GHz 7448 PowerPC CPUs. In order to implement an optimal software architecture on which a distributed parallel processing is possible, we measured and analyzed the data delivery time between the CPUs. And the performance of the present algorithm was verified, comparing classification results using the real off-road images acquired under various environmental conditions in conformity with applied classifiers and features. Experiments show the robustness of the classification results on any environmental condition.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.6
• /
• pp.29-35
• /
• 2021

In this paper, we propose a method that combines KNN(K-Nearest Neighbor), Local Map Classification and Bayes Filter as a way to increase the accuracy of location positioning. First, in this technique, Local Map Classification divides the actual map into several clusters, and then classifies the clusters by KNN. And posterior probability is calculated through the probability of each cluster acquired by Bayes Filter. With this posterior probability, the cluster where the robot is located is searched. For performance evaluation, the results of location positioning obtained by applying KNN, Local Map Classification, and Bayes Filter were analyzed. As a result of the analysis, it was confirmed that even if the RSSI signal changes, the location information is fixed to one cluster, and the accuracy of location positioning increases.

• Journal of Internet Computing and Services
• /
• v.22 no.3
• /
• pp.9-16
• /
• 2021

This study aims to develop a human activity recognition (HAR) system as a Deep-Learning (DL) classification model, distinguishing various human activities. We solely rely on the signals from a wristband accelerometer worn by a person for the user's convenience. 3-axis sequential acceleration signal data are gathered within a predefined time-window-slice, and they are used as input to the classification system. We are particularly interested in developing a Deep-Learning model that can outperform conventional machine learning classification performance. A total of 13 activities based on the laboratory experiments' data are used for the initial performance comparison. We have improved classification performance using the Convolutional Neural Network (CNN) combined with an auto-encoder feature reduction and parameter tuning. With various publically available HAR datasets, we could also achieve significant improvement in HAR classification. Our CNN model is also compared against Recurrent-Neural-Network(RNN) with Long Short-Term Memory(LSTM) to demonstrate its superiority. Noticeably, our model could distinguish both general activities and near-identical activities such as sitting down on the chair and floor, with almost perfect classification accuracy.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2008.03a
• /
• pp.1466-1483
• /
• 2008

This paper discusses the development of an up-to-date computerized CPT (Cone Penetration Test) based soil engineering classification system to provide geotechnical engineers with a handy tool for their daily design activities. Five CPT soil engineering classification systems are incorporated in this effort. They include the probabilistic region estimation and fuzzy classification methods, both developed by Zhang and Tumay, the Schmertmann, the Douglas and Olsen, and the Robertson et al. methods. In the probabilistic region estimation method, a conformal transformation is used to determine the soil classification index, U, from CPT cone tip resistance and friction ratio. A statistical correlation is established between U and the compositional soil type given by the Unified Soil Classification System (USCS). The soil classification index, U, provides a soil profile over depth with the probability of belonging to different soil types, which more realistically and continuously reflects the in-situ soil characterization, which includes the spatial variation of soil types. The CPT fuzzy classification on the other hand emphasizes the certainty of soil behavior. The advantage of combining these two classification methods is realized through implementing them into visual basic software with three other CPT soil classification methods for friendly use by geotechnical engineers. Three sites in Louisiana were selected for this study. For each site, CPT tests and the corresponding soil boring results were correlated. The soil classification results obtained using the probabilistic region estimation and fuzzy classification methods are cross-correlated with conventional soil classification from borings logs and three other established CPT soil classification methods.

• The Journal of the Korea Contents Association
• /
• v.8 no.7
• /
• pp.66-74
• /
• 2008

Criticality prediction models to identify most fault-prone modules in the system early in the software development process help in allocation of resources and foster software quality improvement. Many models for identifying fault-prone modules using design complexity metrics have been suggested, but most of them are training models that need training data set. Most organizations cannot use these models because very few organizations have their own training data. This paper builds a prediction model based on a well-known supervised learning model, error backpropagation neural net, using design metrics quantifying SDL system specifications. To solve the problem of other models, this model is trained by generated virtual training data set. Some simulation studies have been performed to investigate feasibility of this model, and the results show that suggested model can be an alternative for the organizations without real training data to predict their software qualities.

• Journal of Information Processing Systems
• /
• v.19 no.6
• /
• 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.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.7
• /
• pp.27-37
• /
• 2020

In this paper, we propose the methodology based on data-driven approach using Natural Language Processing and Machine Learning for classifying requirements into functional requirements and non-functional requirements. Through the analysis of the results of the requirements classification, we have learned that the trained models derived from requirements classification with data-preprocessing and classification algorithm based on the characteristics and information of existing requirements that used term weights based on TF and IDF outperformed the results that used stemming and stop words to classify the requirements into functional and non-functional requirements. This observation also shows that the term weight calculated without removal of the stemming and stop words influenced the results positively. Furthermore, we investigate an optimized method for the study of classifying software requirements into functional and non-functional requirements.

• Journal of Digital Convergence
• /
• v.13 no.12
• /
• pp.187-192
• /
• 2015

If the software is developed to analyze the speech disorder, the application of various converged areas will be very high. This paper implements the user-friendly program based on CART(Classification and regression trees) analysis to distinguish between normal and pathological voices utilizing combination of the acoustical and HOS(Higher-order statistics) parameters. It means convergence between medical information and signal processing. Then the acoustical parameters are Jitter(%) and Shimmer(%). The proposed HOS parameters are means and variances of skewness(MOS and VOS) and kurtosis(MOK and VOK). Database consist of 53 normal and 173 pathological voices distributed by Kay Elemetrics. When the acoustical and proposed parameters together are used to generate the decision tree, the average accuracy is 83.11%. Finally, we developed a program with more user-friendly interface and frameworks.