• Journal of KIISE
• /
• v.45 no.2
• /
• pp.157-164
• /
• 2018

Due to the revolutionary advances in deep learning, performance of pattern recognition has increased significantly in many applications like speech recognition and image recognition, and some systems outperform human-level intelligence in specific domains. Unlike pattern recognition, in this paper, we focus on generating Korean sentences based on a few Korean sentences. We apply variational recurrent auto-encoder (VRAE) and modify the model considering some characteristics of Korean sentences. To reduce the number of words in the model, we apply a word spacing model. Also, there are many Korean sentences which have the same meaning but different word order, even without subjects or objects; therefore we change the unidirectional encoder of VRAE into a bidirectional encoder. In addition, we apply an interpolation method on the encoded vectors from the given sentences, so that we can generate new sentences which are similar to the given sentences. In experiments, we confirm that our proposed method generates better sentences which are semantically more similar to the given sentences.

• Journal of Software Assessment and Valuation
• /
• v.17 no.2
• /
• pp.47-57
• /
• 2021

Neuromorphic technology is proposed to complement the shortcomings of existing artificial intelligence technology by mimicking the human brain structure and computational process with hardware. NA-IDE has also been proposed for developing neuromorphic hardware-based IoT applications. To implement an SNN model in NA-IDE, commonly used input data must be transformed for use in the SNN model. In this paper, we implemented a neural coding method encoder component that converts image data into a spike train signal and uses it as an SNN input. The decoder component is implemented to convert the output back to image data when the SNN model generates a spike train signal. If the decoder component uses the same parameters as the encoding process, it can generate static data similar to the original data. It can be used in fields such as image-to-image and speech-to-speech to transform and regenerate input data using the proposed encoder and decoder.

• The KIPS Transactions:PartA
• /
• v.12A no.7 s.97
• /
• pp.589-596
• /
• 2005

Source code quality is very Important that software embedded into product is difficult to change. In order to improve source code quality, it should be considered the quality of analysis and design models as well as the quality of source code. In this paper, we suggest 'Practice Pattern' as one of practical techniques to improve embedded software source code quality. Practice pattern is a procedural pattern to guide modeling and coding activities in software development phases. We believe that applying our pattern provides the improvement of optimum performance, modularization, and portability for embedded software source code.

• Journal of Korean Society of Rural Planning
• /
• v.20 no.4
• /
• pp.221-232
• /
• 2014

초고령화된 한국 농촌에서 다양한 교육 및 직업 배경을 가진 중장년층 귀농인구의 유입은 고령화 문제에 대한 대안의 하나로 부각되고 있다. 본 연구는 농가 고령화에 대한 귀농인구의 잠재적인 기여도 측면에서 귀농인구의 지역별 유입 유출 패턴을 분석하였다. 분석에는 로그선형모형과 총합레퍼런스코딩을 사용하였고, 분석자료는 통계청의 2013년 귀농 통계이다. 분석결과에 의하면, 귀농인의 절반 정도가 수도권에서 비수도권 지역으로 이주한 인구이고, 이들을 제외하면 귀농인의 대부분은 원 거주지가 있던 도 내에서 이동하고 있다. 귀농인의 귀농 전 대비 귀농 후 지역 내 오즈비(odds ratio)는 지역별로 차이가 있으며, 귀농인의 성별과 연령에 따라서도 지역별 유입 유출패턴이 다르다. 이는 귀농인 유입의 긍정적 효과를 높이기 위해서는 지자체별 차이를 반영한 특화된 정책이 필요하다는 것을 의미한다.

• Journal of the Korean Society for information Management
• /
• v.34 no.1
• /
• pp.241-262
• /
• 2017

This study aims to identify and analyze the Twitter use of domestic public libraries. In order to identify the detailed patterns of Twitter use in library and information services, a content analysis was conducted for the 3,038 tweet data from the top 14 public libraries' accounts on Twitter use. Inductive approach was adopted to develop a coding scheme and open coding was conducted with the entire tweet. Additionally, correspondence analysis was conducted for the result of content analysis to identify how library accounts correspond to specific types. As a result, 3 main categories and 9 sub-categories of public libraries' Twitter use were developed. And the 37 detailed patterns of public libraries' use of Twitter were identified. The identified patterns can provide the libraries interested in Twitter use with guidelines.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.5 no.6
• /
• pp.515-520
• /
• 2004

Core-based system-on-a-chip (SoC) designs present a number of test challenges. Two major problems that are becoming increasingly important are long application time during manufacturing test and high volume of test data. Highly efficient compression techniques have been proposed to reduce storage and application time for high volume data by exploiting the repetitive nature of test vectors. This paper proposes a new test data compression technique for SoC testing. In the proposed technique, compression is achieved by partitioning the test vector set and removing repeating segment. This process has $O(n^{-2})$ time complexity for compression with a simple hardware decoding circuitry. It is shown that the efficiency of the proposed compression technique is comparable with sophisticated software compression techniques with the advantage of easy and fast decoding.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.6
• /
• pp.11-16
• /
• 2017

Multi-level holographic data storage requires modulation codes for avoiding two dimensional inter-symbol interference (2D-ISI). Modulation codes can remove the fatal ISI pattern of neighboring the largest and the smallest symbols. In this paper, we propose a 4-level 4/6 modulation code and its trellis encoding for error correction. The proposed 4/6 modulation code prevents that the symbol 0 and 3 are not adjacent in any direction. Also, we compare the proposed modulation code with the same code rate modulation codes for four-level holographic data storage.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.12 no.7
• /
• pp.1067-1075
• /
• 2001

The fields at the aperture of conical horn antenna with corrugations parallel to the axis have been analyzed using FDTD(Finite Difference Time Domain). Easy calculation depending on the change of the structure of antenna and time reduction can be achieved by 2-D FDTD coding with the first-order Mur ABC(absorbing boundary condition). It is confirmed that the corrugation can reduce phase difference of field on aperture. also it is investigated that the directivity is increased by 6.1 %, 12.9%, and 28.4% with one corrugation, two corrugations, three corrugations, respectively. It is also found that the improvement of the characteristics of the antenna is not proportional to the number of the corrugation but more dependent on the location of the corrugation near the aperture than that far the aperture.

• The Journal of the Korea Contents Association
• /
• v.13 no.5
• /
• pp.37-47
• /
• 2013

The objective of this research is to discover the rule of gameplay related to the task interdependence to analyse the behavior pattern of social gameplay. Previous literatures related to the gameplay were reviewed and game which was suitable for the gameplay of the task interdependence was selected. A party-play includes a team of five people in the experiment during the gameplay with think-aloud method and video/audio data about action protocol and verbal report were collected. The video observation and protocol analysis were conducted to analyse data. The objective coding scheme were developed from consolidated sequence model task analysis. The player's behavior was analysed. The result was revealed that four rules and four modified rules were included into the total eight behavior pattern. A behavior graph integrated with five gameplay was written. The excellent cooperative spot and error and failure place could be identified. The social gameplay behavior graph is expected to be the key practical design guideline on whether the level design and balance design are proper.

• Smart Media Journal
• /
• v.7 no.4
• /
• pp.17-23
• /
• 2018

We proposed a LSTM-based RNN model that can effectively perform the automatic spacing characteristics. For those long or noisy sentences which are known to be difficult to handle within Neural Network Learning, we defined a proper input data format and decoding data format, and added dropout, bidirectional multi-layer LSTM, layer normalization, and attention mechanism to improve the performance. Despite of the fact that Sejong corpus contains some spacing errors, a noise-robust learning model developed in this study with no overfitting through a dropout method helped training and returned meaningful results of Korean word spacing and its patterns. The experimental results showed that the performance of LSTM sequence-to-sequence model is 0.94 in F1-measure, which is better than the rule-based deep-learning method of GRU-CRF.