• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권12호
• /
• pp.4275-4291
• /
• 2021

With the unprecedented growth of textual information on the Internet, an efficient automatic summarization system has become an urgent need. Recently, the neural network models based on the encoder-decoder with an attention mechanism have demonstrated powerful capabilities in the sentence summarization task. However, for paragraphs or longer document summarization, these models fail to mine the core information in the input text, which leads to information loss and repetitions. In this paper, we propose an abstractive document summarization method by applying guidance signals of key sentences to the encoder based on the hierarchical encoder-decoder architecture, denoted as KI-HABS. Specifically, we first train an extractor to extract key sentences in the input document by the hierarchical bidirectional GRU. Then, we encode the key sentences to the key information representation in the sentence level. Finally, we adopt key information representation guided selective encoding strategies to filter source information, which establishes a connection between the key sentences and the document. We use the CNN/Daily Mail and Gigaword datasets to evaluate our model. The experimental results demonstrate that our method generates more informative and concise summaries, achieving better performance than the competitive models.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권7호
• /
• pp.3272-3285
• /
• 2016

Recently, Vector of Locally Aggregated Descriptors (VLAD) has been proposed to index image by compact representations, which encodes powerful local descriptors and makes significant improvement on search performance with less memory compared against the state of art. However, its performance relies heavily on the size of the codebook which is used to generate VLAD representation. It indicates better accuracy needs higher dimensional representation. Thus, more memory overhead is needed. In this paper, we enhance VLAD image representation by using two level hierarchical-codebooks. It can provide more accurate search performance while keeping the VLAD size unchanged. In addition, hierarchical-codebooks are used to construct multiple inverted files for more accurate non-exhaustive search. Experimental results show that our method can make significant improvement on both VLAD image representation and non-exhaustive search.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권1호
• /
• pp.254-276
• /
• 2019

Since the amount of information on the internet is growing rapidly, it is not easy for a user to find relevant information for his/her query. To tackle this issue, the researchers are paying much attention to Document Summarization. The key point in any successful document summarizer is a good document representation. The traditional approaches based on word overlapping mostly fail to produce that kind of representation. Word embedding has shown good performance allowing words to match on a semantic level. Naively concatenating word embeddings makes common words dominant which in turn diminish the representation quality. In this paper, we employ word embeddings to improve the weighting schemes for calculating the Latent Semantic Analysis input matrix. Two embedding-based weighting schemes are proposed and then combined to calculate the values of this matrix. They are modified versions of the augment weight and the entropy frequency that combine the strength of traditional weighting schemes and word embedding. The proposed approach is evaluated on three English datasets, DUC 2002, DUC 2004 and Multilingual 2015 Single-document Summarization. Experimental results on the three datasets show that the proposed model achieved competitive performance compared to the state-of-the-art leading to a conclusion that it provides a better document representation and a better document summary as a result.

• ETRI Journal
• /
• 제44권5호
• /
• pp.794-804
• /
• 2022

Transformer models have had a great impact on natural language processing (NLP) in recent years by realizing outstanding and efficient contextualized language models. Recent studies have used transformer-based language models for various NLP tasks, including Persian named entity recognition (NER). However, in complex tasks, for example, NER, it is difficult to determine which contextualized embedding will produce the best representation for the tasks. Considering the lack of comparative studies to investigate the use of different contextualized pretrained models with sequence modeling classifiers, we conducted a comparative study about using different classifiers and embedding models. In this paper, we use different transformer-based language models tuned with different classifiers, and we evaluate these models on the Persian NER task. We perform a comparative analysis to assess the impact of text representation and text classification methods on Persian NER performance. We train and evaluate the models on three different Persian NER datasets, that is, MoNa, Peyma, and Arman. Experimental results demonstrate that XLM-R with a linear layer and conditional random field (CRF) layer exhibited the best performance. This model achieved phrase-based F-measures of 70.04, 86.37, and 79.25 and word-based F scores of 78, 84.02, and 89.73 on the MoNa, Peyma, and Arman datasets, respectively. These results represent state-of-the-art performance on the Persian NER task.

• 한국통신학회논문지
• /
• 제22권2호
• /
• pp.241-249
• /
• 1997

본 논문은 통신 프로토콜의 적합성 시험시 외부 동기와 입출력 시험 항목의 비용이 고려되어야 할 분산 시험 환경에서 적용될 수 있는 동기적 시험 항목 생성을 위한 새로운 방법을 제시한다. 이 방법은 유한 상태 기계로 표현된 프로토콜의 명세로 부터 이중 방향 그래프를 구성한 후 최소 길이의 동기적 시험 항목을 생성하기 위해 rural Chinese postman tour(RCPT) 알고리듬을 적용한다. 또한 시험시 주어진 유한 상태 기계의 각 상태에 대한 상태 식별 항목으로써 synchronizable distinguishing sequence(SDS)들을 사용한다. 본 논문의 방법은 원거리 시험기들을 사용하는 프로토콜 시험 구조에서 기 결정된 시험 항목을 적용할 때 시험 대상과 이들 시험기들간에 발생할 수 있는 동기화 문제에 대한 해결책으로 제시될 수 있다.

• 전기의세계
• /
• 제21권6호
• /
• pp.9-16
• /
• 1972

This work intends to study the machine dynamics in the state-space approah and to formulate the operating characteristics of a namplidyne generator, with balanced control field winding and an auxiliary feedback winding for compensating purpose. In the derivation of the dynamic equation, investigations on the demagnetization effects are accentuated, based on the magnetic interlinks between windings of the machine. From the machine dynamic relation obtained, a state-variable representation of the machine dynamics is approached in the first part of this work.

• 대한화학회지
• /
• 제57권2호
• /
• pp.172-175
• /
• 2013

We compare the relative usefulness of common basis functions and numerical integration methods in representing complex resonance state encountered in the molecular scattering problem of aluminum dimer electronic predissociation. Specifically, the basis set size and computing CPU times are monitored in order to find the minimum requirement for ensuring the modest accuracy of calculated resonance energies (0.1 $cm^{-1}$) for more than 100 resonance states. The combination of the so-called one-dimensional box eigenfunctions and energy-dependent boundary functions are found to be most efficient if integration is done using the basis set quadrature rules.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 1996년도 추계학술대회논문집; 한국과학기술회관, 8 Nov. 1996
• /
• pp.255-259
• /
• 1996

This paper presents active vibration control of a rotating cantilevered beam using piezoceramic actuators. A governing equation of motion is obtained by the Hamilton's principle and expressed in the state space representation. Subsequently, an H$_{\infty}$ control which is robust to system uncertainties is synthesized through the loop shaping design procedure. Computer simulations for the steady-state vibration control are undertaken in order to demonstrate the effectiveness and robustness of the proposed control methodology..y.

• 한국정밀공학회지
• /
• 제19권5호
• /
• pp.168-175
• /
• 2002

The hydraulic acoustic attenuator fur an automotive active suspension system is so highly nonlinear and of high order that the analysis in time-domain has been performed quite little. In this paper, a state-space representation of the dynamics for a hydraulic attenuator was presented utilizing the electrical analogy. And the results of experiment were compared with those of simulation to validate the state-space model proposed. The comparison revealed that the state-space model proposed is practically applicable to estimate the dynamic responses of the hydraulic attenuator in time-domain.

• Journal of information and communication convergence engineering
• /
• 제6권3호
• /
• pp.310-314
• /
• 2008

This paper proposes a very novel method which makes it possible that state feedback controller can be designed for unknown dynamic system with measurable states. This novel method uses the support vector machines (SVM) with its function approximation property. It works together with RLS (Recursive least-squares) algorithm. The RLS algorithm is used for the identification of input-output relationship. A virtual state space representation is derived from the relationship and the SVM makes the relationship between actual states and virtual states. A state feedback controller can be designed based on the virtual system and the SVM makes the controller with actual states. The results of this paper can give many opportunities that the state feedback control can be applied for unknown dynamic systems.