• The Journal of the Acoustical Society of Korea
• /
• v.22 no.8
• /
• pp.727-739
• /
• 2003

In this paper, we propose a generation method of regression classes for adaptation in the HM-Net (Hidden Markov Network) system. The MLLR (Maximum Likelihood Linear Regression) adaptation approach is applied to the HM-Net speech recognition system for expressing the characteristics of speaker effectively and the use of HM-Net in various tasks. For the state level sharing, the context domain state splitting of PDT-SSS (Phonetic Decision Tree-based Successive State Splitting) algorithm, which has the contextual and time domain clustering, is adopted. In each state of contextual domain, the desired phoneme classes are determined by splitting the context information (classes) including target speaker's speech data. The number of adaptation parameters, such as means and variances, is autonomously controlled by contextual domain state splitting of PDT-SSS, depending on the context information and the amount of adaptation utterances from a new speaker. The experiments are performed to verify the effectiveness of the proposed method on the KLE (The center for Korean Language Engineering) 452 data and YNU (Yeungnam Dniv) 200 data. The experimental results show that the accuracies of phone, word, and sentence recognition system increased by 34∼37%, 9%, and 20%, respectively, Compared with performance according to the length of adaptation utterances, the performance are also significantly improved even in short adaptation utterances. Therefore, we can argue that the proposed regression class method is well applied to HM-Net speech recognition system employing MLLR speaker adaptation.

• Journal of Broadcast Engineering
• /
• v.24 no.3
• /
• pp.495-505
• /
• 2019

Typical algorithms of deep learning include CNN(Convolutional Neural Networks), which are mainly used for image recognition, and RNN(Recurrent Neural Networks), which are used mainly for speech recognition and natural language processing. Among them, CNN is able to learn from filters that generate feature maps with algorithms that automatically learn features from data, making it mainstream with excellent performance in image recognition. Since then, various algorithms such as R-CNN and others have appeared in object detection to improve performance of CNN, and algorithms such as YOLO(You Only Look Once) and SSD(Single Shot Multi-box Detector) have been proposed recently. However, since these deep learning-based detection algorithms determine the success of the detection in the still images, stable object tracking and detection in the video requires separate tracking capabilities. Therefore, this paper proposes a method of combining Kalman filters into deep learning-based detection networks for improved object tracking and detection performance in the video. The detection network used YOLO v2, which is capable of real-time processing, and the proposed method resulted in 7.7% IoU performance improvement over the existing YOLO v2 network and 20 fps processing speed in FHD images.

• Korean Journal of Clinical Pharmacy
• /
• v.23 no.2
• /
• pp.106-113
• /
• 2013

Background: For Korean pharmaceutical industry to continue to grow, it is requisite to enter the global markets of developed countries. However, the export volume has fallen short of 10% of the gross sales and the industry has only recently warming up to prepare the globalization along with suchlike the Columbus Project. Purpose: This research was conducted to identify the difficulties Korean pharmaceutical companies perceived and to discover the gap in the needs for the government aid the companies have been seeking in the purpose of entering the developed pharmaceutical markets. Method: A survey method was used for this research. Six experts were surveyed and provided comments for the pre-questionnaire. Then, a final questionnaire was developed consisting of 10 items on regulatory-related and another 10 items on non-regulatory-related factors in drug exportation using the Likert scale (1 to 5). The survey sample was 30 Korean companies which have participated in the Columbus Project since 2010. Results: Nineteen (63%) companies responded to the survey. Most companies perceived difficulty (mean = 4.19) over the entire pathway of the regulatory process of global markets. Clinical trials and post-marketing surveillance were remarked as the most difficult barrier to follow the regulatory globalization. Among non-regulatory related factors, marketing, arranging a distribution network, obtaining experts, and projecting a timeline in exportation were brought up as the most difficulty. Conclusion: Especially, cost and language barrier were considered as the main cause producing these difficulties across regulatory and non-regulatory processes and accordingly, securing both long term budget and experts at governmental level was suggested by the domestic pharmaceutical companies.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.1
• /
• pp.49-55
• /
• 2011

There are tries to unite through consilience in many fields. Interdisciplinary research is an instance of those. Linguistic studies called linguistic informatics or quantitative linguistics is a field of interdisciplinary research related with statistics linguists have studied chiefly statistics and linguistics. In the statistical aspect, there is need to supplement somewhat of the result of researches by linguists. This study shows statistical method can supplement insufficient objectivity in linguistic studies, and examines the way to raise a degree of completion of interdisciplinary research on statistics and linguistics. This study also shows an introduction and application of the statistical method can be useful for the deduction of objective linguistic information in linguistic studies.

• Journal of KIISE:Software and Applications
• /
• v.37 no.10
• /
• pp.779-787
• /
• 2010

Since robots have been used widely recently, research about human-robot communication is in process actively. Typically, natural language processing or gesture generation have been applied to human-robot interaction. However, existing methods for communication among robot and human have their limits in performing only static communication, thus the method for more natural and realistic interaction is required. In this paper, an emotional gesture based dialogue management system is proposed for sophisticated human-robot communication. The proposed system performs communication by using the Bayesian networks and pattern matching, and generates emotional gestures of robots in real-time while the user communicates with the robot. Through emotional gestures robot can communicate the user more efficiently also realistically. We used behavior networks as the gesture generation method to deal with dialogue situations which change dynamically. Finally, we designed a usability test to confirm the usefulness of the proposed system by comparing with the existing dialogue system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.8
• /
• pp.3168-3186
• /
• 2020

Automatic topic essay generation (TEG) is a controllable text generation task that aims to generate informative, diverse, and topic-consistent essays based on multiple topics. To make the generated essays of high quality, a reasonable method should consider both diversity and topic-consistency. Another essential issue is the intrinsic link of the topics, which contributes to making the essays closely surround the semantics of provided topics. However, it remains challenging for TEG to fill the semantic gap between source topic words and target output, and a more powerful model is needed to capture the semantics of given topics. To this end, we propose a pretraining-based contextual self-attention (PC-SAN) model that is built upon the seq2seq framework. For the encoder of our model, we employ a dynamic weight sum of layers from BERT to fully utilize the semantics of topics, which is of great help to fill the gap and improve the quality of the generated essays. In the decoding phase, we also transform the target-side contextual history information into the query layers to alleviate the lack of context in typical self-attention networks (SANs). Experimental results on large-scale paragraph-level Chinese corpora verify that our model is capable of generating diverse, topic-consistent text and essentially makes improvements as compare to strong baselines. Furthermore, extensive analysis validates the effectiveness of contextual embeddings from BERT and contextual history information in SANs.

• The Journal of the Korea Contents Association
• /
• v.2 no.4
• /
• pp.40-45
• /
• 2002

The network community provides people a new area which is called a cyber area, not physical area. However, today in the cyber area, the negative function has been causing social problem. In fact, it has not been causing by original function. Specifically speaking, main reason why it's leading to the negative function is that teenagers are logging into inappropriate sites (such as: violent, adult sites) to unnecessary sites (such as: e-mails with unknown signs combined with words, inappropriate language used in certain web board). Not only that, private password and ID's are being used by unknown user to accomplish illegal access. If this problem does not get solved through immediate development of technologies method, the cyber crime will increase in short period of time. The purpose of this paper is to understand each analyzed method which can cope with negative internet function, to discuss suitable management in certain web board an to check with fuzzy intercommunication mood.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.5
• /
• pp.797-806
• /
• 1995

The Client-Server model is a computing technique that efficiently uses distributed resources via networks. But, generating an application using Client- Server model requires many different expertises to developers in comparison to that of using the centralized computing method. That is, and application developer must be familiar with network programming and GUI(Graphical User Interface) techniques in addition to conventional programming skills. Accordingly, the time and man power have been issued for building a Client-Server system. To alleviate these problem, the Client-Server applications development tools are needed To meet such a need, we developed a GUI based tool, called Hanuri/C, for generating Client-Server application programs. In comparison to existing Client-Server tools, Hanuri/C is reinforced Multimedia facilities, Hanuri/C is reinforced Multimedia facilities. Hanuri/C is considered as a front-end part of distributed system software. That is, Hanuri/C is a Client-Server tool similar to Forth Generation Language(4GL) supporting multimedia application on top of the Distributed Computing environment(DCE). In this paper, we present the design and implementation method of the middleware part of Hanuri/c.

• Journal of the Korean association of regional geographers
• /
• v.4 no.2
• /
• pp.139-150
• /
• 1998

Recently in connection with the development of network technology, GIS researches shift their focuses to openGIS which enables to share heterogeneous geodata and geoprocessing resource under the environment of LAN and WAN. This study proposes an easy method to develope GUIs which enables openGIS in the intranet environment. In the developing stage of GUI, the most important thing is to develop effective windows programming technique and apply a internet linking technique to developed GUIs. This study shows an internet linking technique and a GUI developing method using OLE automation and ActiveX/COM standards. Using a case study of Taegu City Library, the study constructs a GIS database and develops its GUI which performs map overlaying, referencing attribute tables, drawing up of thematic map, buffer analysis, and statistical calculations. This study remains same further research areas of developing a new advanced technique for openGIS and its GUI using distributed object model of CORBA, Java language, and JDBC data accessing technique.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.3991-4010
• /
• 2021

The Question Similarity Measurement of Chinese Crop Diseases and Insect Pests (QSM-CCD&IP) aims to judge the user's tendency to ask questions regarding input problems. The measurement is the basis of the Agricultural Knowledge Question and Answering (Q & A) system, information retrieval, and other tasks. However, the corpus and measurement methods available in this field have some deficiencies. In addition, error propagation may occur when the word boundary features and local context information are ignored when the general method embeds sentences. Hence, these factors make the task challenging. To solve the above problems and tackle the Question Similarity Measurement task in this work, a corpus on Chinese crop diseases and insect pests(CCDIP), which contains 13 categories, was established. Then, taking the CCDIP as the research object, this study proposes a Chinese agricultural text similarity matching model, namely, the AgrCQS. This model is based on mixed information extraction. Specifically, the hybrid embedding layer can enrich character information and improve the recognition ability of the model on the word boundary. The multi-scale local information can be extracted by multi-core convolutional neural network based on multi-weight (MM-CNN). The self-attention mechanism can enhance the fusion ability of the model on global information. In this research, the performance of the AgrCQS on the CCDIP is verified, and three benchmark datasets, namely, AFQMC, LCQMC, and BQ, are used. The accuracy rates are 93.92%, 74.42%, 86.35%, and 83.05%, respectively, which are higher than that of baseline systems without using any external knowledge. Additionally, the proposed method module can be extracted separately and applied to other models, thus providing reference for related research.