• 지능정보연구
• /
• 제5권1호
• /
• pp.103-123
• /
• 1999

Input filtering as a preprocessing method is so much crucial to get good performance in time series forecasting. There are a few preprocessing methods (i.e. ARMA outputs as time domain filters, and Fourier transform or wavelet transform as time-frequency domain filters) for handling time series. Specially, the time-frequency domain filters describe the fractal structure of financial markets better than the time domain filters due to theoretically additional frequency information. Therefore, we, first of all, try to describe and analyze specially some issues on the effectiveness of different filtering methods from viewpoint of the performance of a neural network based forecasting. And then we discuss about neural network model architecture issues, for example, what type of neural network learning architecture is selected for our time series forecasting, and what input size should be applied to a model. In this study an input selection problem is limited to a size selection of the lagged input variables. To solve this problem, we simulate on analyzing and comparing a few neural networks having different model architecture and also use an embedding dimension measure as chaotic time series analysis or nonlinear dynamic analysis to reduce the dimensionality (i.e. the size of time delayed input variables) of the models. Throughout our study, experiments for integration methods of joint time-frequency analysis and neural network techniques are applied to a case study of daily Korean won / U. S dollar exchange returns and finally we suggest an integration framework for future research from our experimental results.

• 한국경영과학회지
• /
• 제25권1호
• /
• pp.67-84
• /
• 2000

In this paper, we address a network design problem including the decision of hub facility locatiions which is typically found in a communicatio and a transportation network design studies. Due to the administrative and the geographical restrictions, it is common to assume that each user should be assigned to only one hub facility. To construct such a network, three types of network costs should be considered: the fixed costs of establishing the hubs and the arcs in the network, and the variable costs associated with transversing the network. The complex problem is formulated as a mixed IP embedding a multicommodity flow problem. Exploiting its special structure, a dual-based heuristic is then developed, which yields near-optimal design plans. The test results indicate that the heuristic is an effective way to solve this computationally complex problem.

• International Journal of Computer Science & Network Security
• /
• 제21권12호
• /
• pp.21-28
• /
• 2021

In this paper, a novel data transferring method is introduced that can transmit sensor data without using data bandwidth or an extra-processing cycle in a delay insensitive network. The proposed method uses human devices as Mules, does not disturb the device owner for permission, and saves energy while transferring sensor data to the collection hub in a wireless sensor network. This paper uses IP addressing technique as the data transferring mechanism by embedding the sensor data with the IP address of a Mule. The collection hub uses the ARP sequence method to extract the embedded data from the IP address. The proposed method follows WiFi standard in its every step and ends when data collection is over. Every step of the proposed method is discussed in detail with the help of figures in the paper.

• Journal of Information Processing Systems
• /
• 제16권6호
• /
• pp.1250-1260
• /
• 2020

This study proposes a novel deep neural network model that can accurately detect objects and their relationships in an image and represent them as a scene graph. The proposed model utilizes several multimodal features, including linguistic features and visual context features, to accurately detect objects and relationships. In addition, in the proposed model, context features are embedded using graph neural networks to depict the dependencies between two related objects in the context feature vector. This study demonstrates the effectiveness of the proposed model through comparative experiments using the Visual Genome benchmark dataset.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2004년도 가을 학술발표논문집 Vol.31 No.2 (1)
• /
• pp.61-63
• /
• 2004

최근에 하이퍼큐브의 망비용을 개선한 상호연결망 하이퍼-스타가 제안되었다. 본 논문에서는 삼진트리가 하이퍼-스타 연결망 HS(2n, n)에 연장율 1로 임베딩 가능함을 보인다.

• 한국정보통신학회논문지
• /
• 제18권5호
• /
• pp.1116-1121
• /
• 2014

메쉬 구조는 대표적인 상호연결망 중 하나로, VLSI 회로 설계 같은 분야에서 많이 이용되고 있다. 이러한 메쉬 구조에서 지름과 고장허용도를 개선한 연결망으로 토러스와 하이퍼-토러스 연결망이 있다. 본 논문에서는 토러스 구조 T(4k,2l)와 하이퍼-토러스 네트워크 QT(m,n) 사이의 임베딩을 분석한다. 토러스 T(4k,2l)는 QT(m,n)에 연장율 5, 밀집율 4, 확장율 1에 임베딩 가능하고, QT(m,n)은 T(4k,2l)에 연장율 3, 밀집율 3, 확장율 1에 임베딩 가능함을 보인다.

• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.830-841
• /
• 2023

Tokenization is the process of segmenting the input text into smaller units of text, and it is a preprocessing task that is mainly performed to improve the efficiency of the machine learning process. Various tokenization methods have been proposed for application in the field of natural language processing, but studies have primarily focused on efficiently segmenting text. Few studies have been conducted on the Korean language to explore what tokenization methods are suitable for document classification task. In this paper, an exploratory study was performed to find the most suitable tokenization method to improve the performance of a representative spatio-temporal document classifier in Korean. For the experiment, a convolutional neural network model was used, and for the final performance comparison, tasks were selected for document classification where performance largely depends on the tokenization method. As a tokenization method for comparative experiments, commonly used Jamo, Character, and Word units were adopted. As a result of the experiment, it was confirmed that the tokenization of word units showed excellent performance in the case of representative spatio-temporal document classification task where the semantic embedding ability of the token itself is important.

• 컴퓨터교육학회논문지
• /
• 제13권5호
• /
• pp.91-102
• /
• 2010

스타 그래프는 노드 대칭성, 최대 고장 허용도, 계층적 분할 성질을 갖고, 하이퍼큐브보다 망 비용이 개선된 널리 알려진 상호 연결망이다. 본 연구에서는 스타 그래프와 그의 변형된 그래프들 상호 간의 임베딩 방법을 제안한다. 버블정렬 그래프가 팬케익 그래프와 스타 그래프에 각각 연장율 3, 확장율 1로 임베딩 가능함을 보이고, 팬케익 그래프가 버블정렬그래프에 임베딩 하는 연장율 비용이 O($n^2$)임을 보인다. 그리고 스타 그래프가 팬케익 그래프에 연장율 4, 확장율 1로 임베딩 가능함을 보인다. 또한 스타그래프를 버블정렬 그래프에, 팬케익 그래프를 스타 그래프에 임베딩 하는 연장율 비용이 각각 O(n)임을 보인다.

• 한국멀티미디어학회논문지
• /
• 제24권10호
• /
• pp.1380-1390
• /
• 2021

The video face recognition (FR) is one of the most popular researches in the field of computer vision due to a variety of applications. In particular, research using the attention mechanism is being actively conducted. In video face recognition, attention represents where to focus on by using the input value of the whole or a specific region, or which frame to focus on when there are many frames. In this paper, we propose a novel attention based deep learning method. Main novelties of our method are (1) the use of combining two loss functions, namely weighted Softmax loss function and a Triplet loss function and (2) the feasibility of end-to-end learning which includes the feature embedding network and attention weight computation. The feature embedding network has a positive effect on the attention weight computation by using combined loss function and end-to-end learning. To demonstrate the effectiveness of our proposed method, extensive and comparative experiments have been carried out to evaluate our method on IJB-A dataset with their standard evaluation protocols. Our proposed method represented better or comparable recognition rate compared to other state-of-the-art video FR methods.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2019년도 제31회 한글 및 한국어 정보처리 학술대회
• /
• pp.272-275
• /
• 2019

단어 중의성 해소 방법은 지식 정보를 활용하여 문제를 해결하는 지식 기반 방법과 각종 기계학습 모델을 이용하여 문제를 해결하는 지도학습 방법이 있다. 지도학습 방법은 높은 성능을 보이지만 대량의 정제된 학습 데이터가 필요하다. 반대로 지식 기반 방법은 대량의 정제된 학습데이터는 필요없지만 높은 성능을 기대할수 없다. 최근에는 이러한 문제를 보완하기 위해 지식내에 있는 정보와 정제된 학습데이터를 기계학습 모델에 학습하여 단어 중의성 해소 방법을 해결하고 있다. 가장 많이 활용하고 있는 지식 정보는 상위어(Hypernym)와 하위어(Hyponym), 동의어(Synonym)가 가지는 의미설명(Gloss)정보이다. 이 정보의 표상을 기존의 문장의 표상과 같이 활용하여 중의성 단어가 가지는 의미를 파악한다. 하지만 정확한 문장의 표상을 얻기 위해서는 단어의 표상을 잘 만들어줘야 하는데 기존의 방법론들은 모두 문장내의 문맥정보만을 파악하여 표현하였기 때문에 정확한 의미를 반영하는데 한계가 있었다. 본 논문에서는 의미정보와 문맥정보를 담은 단어의 표상정보를 만들기 위해 구문정보, 의미관계 그래프정보를 GCN(Graph Convolutional Network)를 활용하여 임베딩을 표현하였고, 기존의 모델에 반영하여 문맥정보만을 활용한 단어 표상보다 높은 성능을 보였다.