• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.993-996
• /
• 2021

챗봇의 사용 용도는 일상 대화와 소비자 응대를 넘어서 심리 상담 용도로 확장하고 있다. 이 연구에서는 챗봇-사람 채팅에서 무작위로 추출한 '우울'과 관련된 대화 데이터를 텍스트마이닝 기법으로 분석하여 채팅에서의 우울 관련 담론 주제를 파악하였다. 더불어 정성 분석을 통해 사용자들이 챗봇에 털어놓고 있는 '우울' 의 종류를 범주화하고 분류하여, 트위터의 '우울' 데이터와의 차이점을 비교하였다. 이를 통해 챗봇 데이터의 '우울' 대화만의 특징을 파악하고, 우울 증상 탐지와 그에 따른 적절한 심리지원 정보를 제공하는 서비스 디자인의 착안점을 제시한다.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.183-203
• /
• 2018

News articles are the most suitable medium for examining the events occurring at home and abroad. Especially, as the development of information and communication technology has brought various kinds of online news media, the news about the events occurring in society has increased greatly. So automatically summarizing key events from massive amounts of news data will help users to look at many of the events at a glance. In addition, if we build and provide an event network based on the relevance of events, it will be able to greatly help the reader in understanding the current events. In this study, we propose a method for extracting event networks from large news text data. To this end, we first collected Korean political and social articles from March 2016 to March 2017, and integrated the synonyms by leaving only meaningful words through preprocessing using NPMI and Word2Vec. Latent Dirichlet allocation (LDA) topic modeling was used to calculate the subject distribution by date and to find the peak of the subject distribution and to detect the event. A total of 32 topics were extracted from the topic modeling, and the point of occurrence of the event was deduced by looking at the point at which each subject distribution surged. As a result, a total of 85 events were detected, but the final 16 events were filtered and presented using the Gaussian smoothing technique. We also calculated the relevance score between events detected to construct the event network. Using the cosine coefficient between the co-occurred events, we calculated the relevance between the events and connected the events to construct the event network. Finally, we set up the event network by setting each event to each vertex and the relevance score between events to the vertices connecting the vertices. The event network constructed in our methods helped us to sort out major events in the political and social fields in Korea that occurred in the last one year in chronological order and at the same time identify which events are related to certain events. Our approach differs from existing event detection methods in that LDA topic modeling makes it possible to easily analyze large amounts of data and to identify the relevance of events that were difficult to detect in existing event detection. We applied various text mining techniques and Word2vec technique in the text preprocessing to improve the accuracy of the extraction of proper nouns and synthetic nouns, which have been difficult in analyzing existing Korean texts, can be found. In this study, the detection and network configuration techniques of the event have the following advantages in practical application. First, LDA topic modeling, which is unsupervised learning, can easily analyze subject and topic words and distribution from huge amount of data. Also, by using the date information of the collected news articles, it is possible to express the distribution by topic in a time series. Second, we can find out the connection of events in the form of present and summarized form by calculating relevance score and constructing event network by using simultaneous occurrence of topics that are difficult to grasp in existing event detection. It can be seen from the fact that the inter-event relevance-based event network proposed in this study was actually constructed in order of occurrence time. It is also possible to identify what happened as a starting point for a series of events through the event network. The limitation of this study is that the characteristics of LDA topic modeling have different results according to the initial parameters and the number of subjects, and the subject and event name of the analysis result should be given by the subjective judgment of the researcher. Also, since each topic is assumed to be exclusive and independent, it does not take into account the relevance between themes. Subsequent studies need to calculate the relevance between events that are not covered in this study or those that belong to the same subject.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.2
• /
• pp.295-310
• /
• 2014

DCS (Distributed Control System), the main control system of power plants, is an automated system for enhancing operational efficiency by monitoring, tuning and real-time operation. DCS is becoming more intelligent and open systems as Information technology are evolving. In addition, there are a large amount of investment to enable proactive facility management, maintenance and risk management through the predictive diagnostics. However, new upcoming weaponized malware, such as Stuxnet designed for disrupting industrial control system(ICS), become new threat to the main control system of the power plant. Even though these systems are not connected with any other outside network. The main control systems used in the power plant usually have been used for more than 10 years. Also, this system requires the extremely high availability (rapid recovery and low failure frequency). Therefore, installing updates including security patches is not easy. Even more, in some cases, installing security updates can break the warranty by the vendor's policy. If DCS is exposed a potential vulnerability, serious concerns are to be expected. In this paper, we conduct the penetration test by using NESSUS, a general-purpose vulnerability scanner under the simulated environment configured with the Ovation version 1.5. From this result, we suggest a log analysis method to detect the security infringement and react the incident effectively.

• The Korean Journal of Zoology
• /
• v.37 no.4
• /
• pp.554-564
• /
• 1994

세균이 세포내 공생하는 xD strain과 모 세포주인 tD strain Amoeba proteus의 열충격 대응의 차이를 알아 보기 위하여 방사선 동위원소로 표지된 아미노산을 Ca2+_less Chalkley's 용액에서 음작용 경로를 통하여 90분 동안 흡수하게 하고, 저온 및 고온 스트레스에 대하여 새로 합성되는 스트레스 대응 단백질의 양상을 1, 2차원 전기영동 및 자기방사 사진법에 의해서 비교하였다 저온(10"C) 충격에 대응하여 아메바는 두 strain 모두 56.0 kDa, pl 6.0 단백질을 강하게 발현하였으며, xD strain에서는 tD strain과 달리 저온 충격 초기에 66 0 kDa, pl 5.5 단백질의 발현이 중단되었다. 한편 고온(33"C) 열충격에 대하여 두 strain 아메바에서 모두 10여종의 단백질이 새합성되는 것으로 확인되었으며, tD 아메바에는 이들 단백질의 새합성이 완만하게 이루어지는데 비하여 xD 아메바에서는 그중 66.0 kDa 단백질이 고온 대응 단백질로서 신속하게 새합성되는 것으로 나타났다. 이외에도 2차원 전기 영동 분석을 통하여 열충격에 의해서 발현이 촉진되는 다수의 단백질들을 탐지하였다 탐지된 아메바의 열충격 단백질은 분자량에 따라 hsp100군 2종, hsp90군, 3종, hsp70군 및 hsp60군 각 1종, 그리고 small csp군 4종으로 분류해 볼 수 있었다 두 분석의 결과를 종합해 보면 tD 아메바에는 저온 및 고온 충격에 대하여 열충격 단백질의 합성이 완만하게 상승하는 데 비하여 xD strain에서는 신속하게 이루어졌다. 이상의 결과로 보아 아메바의 세포내 공생 세균은 숙주의 열충격 대응기작에 변화를 야기한 것으로 판단된다한 것으로 판단된다. 10mg과 20mg의 estrogen 처리구 사이에 유두 직경, 길이 그리고 용적의 증가량에 있어서는 차이가 없었다. 10mg 및 20mg의 estrogen 처리는 초발정일령을 각각 20일 및 124일 단축시켰다. 전체적으로 이러한 결과는 송아지에 estradiol의 삽입은 성장과 유선 발달을 촉진시키고 초발정일령을 단축시킬수 있다는 것을 강력하게 지적한다. 일치하지 않으므로 더욱 정밀한 조사를 실시하여 분류학상의 위치를 정확히 밝혀 볼 필요가 있을 것으로 생각되었다.연한 도구이자 정신활동으로 보게함으로써, 주제 및 연구방법에서 획일성보다 다양성과 창조성이 강조되고 있다. 그리고 연구에 있어서 주제 의 다양성을 통해 보다 현실생활에 밀접하게 연결되어야 할 필요성은 학문이나 과학의 사회 성에 대한 새로운 인식을 가져다 주고 있다. 이러한 지리교육과정의 좌표의 변화된 측면들 을 고려하여, 지리교육과정의 새로운 방향은 다음의 세가지로 모색될 수 있다. 첫째, 爭點中 心 地理敎育課程이다. 사회쟁점에 대한 접근은 쟁점의 이해와 문제해결에의 지리적 관점의 활용을 통해 학습내용의 시사성과 사실성을 높힐 수 있다. 이때 문제해결능력을 통해 현대 시민의 자질 및 능력을 기를 수 있음은 물론, 다른 한편으로 실제세계 즉 학생의 실생활, 사 회, 국가, 세계에서 일어나는 일들과의 관련성을 갖게 함으로써, 내적 동기화와 외적인 자극 을 강력하게 결합할 수 있을 것이다. 이는 개인적 유관적합성과 사회적 유관적합성을 동시 에 확보하는 데 유리할 것이다. 둘째, 思考中心 地理敎育課程이다. 지리교육은 학생들을 지 식 및 기능의 숙달자가 되도록 할 것이 아니라 기본적 문장해독력의 수준을 넘어 능력있는 사고자로 길러내는 것을 목표로 하여야 한다.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.11
• /
• pp.273-280
• /
• 2020

Automatic Static Analysis Tools help developers to quickly find potential defects in source code with less effort. However, the tools reports a large number of false positive warnings which do not have to fix. In our study, we proposed an artificial neural network-based warning classification method using topic models of source code blocks. We collect revisions for fixing bugs from software change management (SCM) system and extract code blocks modified by developers. In deep learning stage, topic distribution values of the code blocks and the binary data that present the warning removal in the blocks are used as input and target data in an simple artificial neural network, respectively. In our experimental results, our warning classification model based on neural network shows very high performance to predict label of warnings such as true or false positive.

• Journal of Korea Multimedia Society
• /
• v.6 no.3
• /
• pp.409-418
• /
• 2003

Object tracking in video is one of subject that computer vision and several practical application field have interest in several years. This paper proposes real time object tracking and face region extraction method that can be applied to security and supervisory system field. For this, in limited environment that camera is fixed and there is seldom change of background image, proposed method detects position of object and traces motion using difference between input image and background image. The system creates adaptive background image and extracts pixels in object using line scan method for more stable object extraction. The real time object tracking is possible through establishment of MBR(Minimum Bounding Rectangle) using extracted pixels. Also, effectiveness for security and supervisory system is improved due to extract face region in established MBR. And through an experiment, the system shows fast real time object tracking under limited environment.

• Spatial Information Research
• /
• v.16 no.4
• /
• pp.479-488
• /
• 2008

This paper performs investigation of the state-of-the-art approaches to building extraction in terms of their sensory input data and methodologies. For the last decades, there have been many types of sensory input data introduced into the mapping science and engineering field, which are considerably diverse in aspects of spatial resolution and data processing. With the cutting-edge technology in this field, accordingly, one of the key issues in GIS is to reconstruct three -dimensional virtual models of the real world to meet the requirements occurring in spatial applications such as urban design, disaster management, and civil works. Thus, this study investigates the strengths and weaknesses of previous approaches to automating building extraction with two categories - building detection and modeling and with sensor types categorized. The findings in this study can be utilized in enhancing automation algorithms and choosing suitable sensors, so that they can be optimized for a specific purpose.

• The KIPS Transactions:PartD
• /
• v.18D no.6
• /
• pp.493-500
• /
• 2011

The trends of technical terms express the changes of particular subjects in a specific research field over time. However, the amount of academic literature and patent data is too large to be analyzed by human resources. In this paper, we propose a method that can detect and analyze the trends of terms by modeling the life cycle of the terms. The proposed method is composed of the following steps. First, the technical terms are extracted from academic literature data, and the TDVs(Term Dominance Values) of terms are computed on a periodic basis. Based on the TDVs, the life cycles of terms are modeled, and technical terms with similar temporal patterns of the life cycles are classified into the same trends class. The experiments shown in this paper is performed by exploiting the NDSL academic literature data maintained by KISTI.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.8C
• /
• pp.675-680
• /
• 2007

This paper presents a new method for classification of underwater transient signals, which employs frame-based decision with Mel Frequency Cepstral Coefficients(MFCC). The MFCC feature vector is extracted frame-by-frame basis for an input signal that is detected as a transient signal, and Euclidean distances are calculated between this and all MFCC feature. vectors in the reference database. Then each frame of the detected input signal is mapped to the class having minimum Euclidean distance in the reference database. Finally the input signal is classified as the class that has maximum mapping rate in the reference database. Experimental results demonstrate that the proposed method is very promising for classification of underwater transient signals.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.7
• /
• pp.275-288
• /
• 2019

In this paper, we consider a rumor source inference problem when sufficiently many nodes heard the rumor in the network. This is an important problem because information spread in networks is fast in many real-world phenomena such as diffusion of a new technology, computer virus/spam infection in the internet, and tweeting and retweeting of popular topics and some of this information is harmful to other nodes. This problem has been much studied, where it has been shown that the detection probability cannot be beyond 31% even for regular trees if the number of infected nodes is sufficiently large. Motivated by this, we study the impact of query that is asking some additional question to the candidate nodes of the source and propose budget assignment algorithms of a query when the network administrator has a finite budget. We perform various simulations for the proposed method and obtain the detection probability that outperforms to the existing prior works.