• Journal of Information Processing Systems
• /
• 제17권3호
• /
• pp.599-614
• /
• 2021

This paper proposes a digital signage system based on an intelligent recommendation model. The proposed system consists of a server and an edge. The server manages the data, learns the advertisement recommendation model, and uses the trained advertisement recommendation model to determine the advertisements to be promoted in real time. The advertisement recommendation model provides predictions for various products and probabilities. The purchase index between the product and weather data was extracted and reflected using correlation analysis to improve the accuracy of predicting the probability of purchasing a product. First, the user information and product information are input to a deep neural network as a vector through an embedding process. With this information, the product candidate group generation model reduces the product candidates that can be purchased by a certain user. The advertisement recommendation model uses a wide and deep recommendation model to derive the recommendation list by predicting the probability of purchase for the selected products. Finally, the most suitable advertisements are selected using the predicted probability of purchase for all the users within the advertisement range. The proposed system does not communicate with the server. Therefore, it determines the advertisements using a model trained at the edge. It can also be applied to digital signage that requires immediate response from several users.

• Journal of Information Science Theory and Practice
• /
• 제10권spc호
• /
• pp.39-45
• /
• 2022

For national industrial development in the Fourth Industrial Revolution, it is necessary to provide researchers with appropriate job information. This can be achieved by interconnecting the National Science and Technology Standard Classification System used for management of research activity with the Korean Employment Classification of Occupations used for job information management. In the present study, an automatic linkage model of classification systems is introduced based on a pre-trained language model for interconnecting science and technology information with job information. We propose for the first time an automatic model for linkage of classification systems. Our model effectively maps similar classes between the National Science & Technology Standard Classification System and Korean Employment Classification of Occupations. Moreover, the model increases interconnection performance by considering hierarchical features of classification systems. Experimental results show that precision and recall of the proposed model are about 0.82 and 0.84, respectively.

• Journal of Mechanical Science and Technology
• /
• 제20권7호
• /
• pp.917-928
• /
• 2006

In recent decades, Artificial Neural Networks (ANNs) have become the focus of considerable attention in many disciplines, including robot control, where they can be used to solve nonlinear control problems. One of these ANNs applications is that of the inverse kinematic problem, which is important in robot path planning. In this paper, a neural network is employed to analyse of inverse kinematics of PUMA 560 type robot. The neural network is designed to find exact kinematics of the robot. The neural network is a feedforward neural network (FNN). The FNN is trained with different types of learning algorithm for designing exact inverse model of the robot. The Unimation PUMA 560 is a robot with six degrees of freedom and rotational joints. Inverse neural network model of the robot is trained with different learning algorithms for finding exact model of the robot. From the simulation results, the proposed neural network has superior performance for modelling complex robot's kinematics.

• 제어로봇시스템학회논문지
• /
• 제4권4호
• /
• pp.551-560
• /
• 1998

This paper presents a neural network model representing complex hydro-thermo-dynamic characteristics of a steam generator in nuclear power plants. The key modeling processes include training data gathering process, analysis of system dynamics and determining of the neural network structure, training process, and the final process for validation of the trained model. In this paper, we suggest a training data gathering method from an unstable steam generator so that the data sufficiently represent the dynamic characteristics of the plant over a wide operating range. In addition, we define the inputs and outputs of neural network model by analyzing the system dimension, relative degree, and inputs/outputs of the plant. Several types of neural networks are applied to the modeling and training process. The trained networks are verified by using a class of test data, and their performances are discussed.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.367-369
• /
• 2022

본 논문에서는 훈련된 모델이 분류에 실패한 이미지들에 대한 정확한 결과를 확인하기 위해 다중 클래스의 이미지 분류를 구현하면서 각각의 클래스에 맞게 레이블 매핑을 하여 예측 결과를 확인했다. Kaggle의 Intel Image Classification 데이터셋을 사용하여 CNN 모델을 구축하고 훈련을 진행하였으며, 테스트 데이터셋의 이미지들을 레이블 매핑을 통해 다중 클래스의 이미지들이 매핑된 레이블 값과 모델이 분류한 값을 비교하였다.

• 로봇학회논문지
• /
• 제19권2호
• /
• pp.196-202
• /
• 2024

The ability to understand given environments and plan a sequence of actions leading to goal state is crucial for personal service robots. With recent advancements in deep learning, numerous studies have proposed methods for state representation in planning. However, previous works lack explicit information about relationships between objects when the state observation is converted to a single visual embedding containing all state information. In this paper, we introduce graph-based state representation that incorporates both object and relationship features. To leverage these advantages in addressing the task planning problem, we propose a Graph Neural Network (GNN)-based subgoal prediction model. This model can extract rich information about object and their interconnected relationships from given state graph. Moreover, a search-based algorithm is integrated with pre-trained subgoal prediction model and state transition module to explore diverse states and find proper sequence of subgoals. The proposed method is trained with synthetic task dataset collected in simulation environment, demonstrating a higher success rate with fewer additional searches compared to baseline methods.

• ETRI Journal
• /
• 제46권3호
• /
• pp.513-525
• /
• 2024

Deep neural networks trained on labeled medical data face major challenges owing to the economic costs of data acquisition through expensive medical imaging devices, expert labor for data annotation, and large datasets to achieve optimal model performance. The heterogeneity of diseases, such as Alzheimer's disease, further complicates deep learning because the test cases may substantially differ from the training data, possibly increasing the rate of false positives. We propose a reconstruction-based self-supervised anomaly detection model to overcome these challenges. It has a dual-subnetwork encoder that enhances feature encoding augmented by skip connections to the decoder for improving the gradient flow. The novel encoder captures local and global features to improve image reconstruction. In addition, we introduce an entropy-based image conversion method. Extensive evaluations show that the proposed model outperforms benchmark models in anomaly detection and classification using an encoder. The supervised and unsupervised models show improved performances when trained with data preprocessed using the proposed image conversion method.

• 정보관리학회지
• /
• 제38권1호
• /
• pp.221-242
• /
• 2021

본 연구에서는 사전학습 된 언어 모델을 기반으로 양방향 게이트 순환 유닛 모델과 조건부 랜덤 필드 모델을 활용하여 참고문헌을 구성하는 메타데이터를 자동으로 인식하기 위한 연구를 진행하였다. 실험 집단은 2018년에 발행된 학술지 40종을 대상으로 수집한 PDF 형식의 학술문헌 53,562건을 규칙 기반으로 분석하여 추출한 참고문헌 161,315개이다. 실험 집합을 구축하기 위하여 PDF 형식의 학술 문헌에서 참고문헌을 분석하여 참고문헌의 메타데이터를 자동으로 추출하는 연구를 함께 진행하였다. 본 연구를 통하여 가장 높은 성능을 나타낸 언어 모델을 파악하였으며 해당 모델을 대상으로 추가 실험을 진행하여 학습 집합의 규모에 따른 인식 성능을 비교하고 마지막으로 메타데이터별 성능을 확인하였다.

• 정보학연구
• /
• 제3권4호
• /
• pp.1-12
• /
• 2000

음성은 특성에 따라 고음성분이 강한 음성과 저음성분이 강한 음성으로 구분할 수 있다. 그러나 이제까지 음성인식의 연구에 있어서는 이러한 특성을 고려하지 않고, 인식기를 구성함으로써 상대적으로 낮은 인식률과 인식모델을 구성할 때 많은 데이터를 필요로 하고 있다. 본 논문에서는 화자의 이러한 특성을 포만트 주파수를 이용하여 구분할 수 있는 방법을 제안하고, 화자음성의 고음과 저음특성을 반영하여 인식모델을 구성한 후 인식하는 방법을 제안한다. 한국어에서 가능한 47개의 모노폰을 이용하여 인식모델을 구성하였으며, 여성과 남성 각각 20명의 음성을 이용하여 인식모델을 학습시켰다. 포만트 주파수를 추출하여 구성한 포만트 주파수 테이불과 피치 정보값을 이용하여 음성의 특성을 구분한 후, 음성특성에 따라 학습된 인식모델을 이용하여 인식을 수행하였다. 본 논문에서 제안한 시스템을 이용하여 실험한 결과 기존의 방법보다 인식률이 향상됨을 보였다.

• 한국컴퓨터정보학회논문지
• /
• 제26권10호
• /
• pp.157-165
• /
• 2021

최근 대량의 텍스트 분석을 위해 딥 러닝(Deep Learning)을 활용하는 연구들이 활발히 수행되고 있으며, 특히 대량의 텍스트에 대한 학습 결과를 특정 도메인 텍스트의 분석에 적용하는 사전 학습 언어 모델(Pre-trained Language Model)이 주목받고 있다. 다양한 사전 학습 언어 모델 중 BERT(Bidirectional Encoder Representations from Transformers) 기반 모델이 가장 널리 활용되고 있으며, 최근에는 BERT의 MLM(Masked Language Model)을 활용한 추가 사전 학습(Further Pre-training)을 통해 분석 성능을 향상시키기 위한 방안이 모색되고 있다. 하지만 전통적인 MLM 방식은 신조어와 같이 새로운 단어가 포함된 문장의 의미를 충분히 명확하게 파악하기 어렵다는 한계를 갖는다. 이에 본 연구에서는 기존의 MLM을 보완하여 신조어에 대해서만 집중적으로 마스킹을 수행하는 신조어 표적 마스킹(NTM: Newly Coined Words Target Masking)을 새롭게 제안한다. 제안 방법론을 적용하여 포털 'N'사의 영화 리뷰 약 70만 건을 분석한 결과, 제안하는 신조어 표적 마스킹이 기존의 무작위 마스킹에 비해 감성 분석의 정확도 측면에서 우수한 성능을 보였다.