• Korean Journal of Artificial Intelligence
• /
• v.5 no.1
• /
• pp.10-17
• /
• 2017

This study is the first attempt to raise face recognition ability through machine learning algorithm and apply to CRM's information gathering, analysis and application. In other words, through face recognition of VIP customer in distribution field, we can proceed more prompt and subdivided customized services. The interest in machine learning, which is used to implement artificial intelligence, has increased, and it has become an age to automate it by using machine learning beyond the way that a person directly models an object recognition process. Among them, Deep Learning is evaluated as an advanced technology that shows amazing performance in various fields, and is applied to various fields of image recognition. Face recognition, which is widely used in real life, has been developed to recognize criminals' faces and catch criminals. In this study, two image analysis models, TF-SLIM and Inception-V3, which are likely to be used for criminal face recognition, were selected, analyzed, and implemented. As an evaluation criterion, the image recognition model was evaluated based on the accuracy of the face recognition program which is already being commercialized. In this experiment, it was evaluated that the recognition accuracy was good when the accuracy of the image classification was more than 90%. A limit of our study which is a way to raise face recognition is left as a further research subjects.

• The KIPS Transactions:PartB
• /
• v.17B no.4
• /
• pp.263-274
• /
• 2010

To provide context-aware music recommendation service, first of all, we need to catch music mood that a user prefers depending on his situation or context. Among various music characteristics, music mood has a close relation with people‘s emotion. Based on this relationship, some researchers have studied on music mood detection, where they manually select a representative segment of music and classify its mood. Although such approaches show good performance on music mood classification, it's difficult to apply them to new music due to the manual intervention. Moreover, it is more difficult to detect music mood because the mood usually varies with time. To cope with these problems, this paper presents an automatic method to classify the music mood. First, a whole music is segmented into several groups that have similar characteristics by structural information. Then, the mood of each segments is detected, where each individual's preference on mood is modelled by regression based on Thayer's two-dimensional mood model. Experimental results show that the proposed method achieves 80% or higher accuracy.

• STI Policy Review
• /
• v.6 no.2
• /
• pp.105-145
• /
• 2015

Science, technology and innovation (STI) is crucially important to eradicating poverty, and making advances in various areas such as agriculture, health, environment, transport, industry, and telecommunications. Therefore, it is vital to the overall socioeconomic development of nations. The indispensable role of STI in the competitive globalized economy led to several attempts to measure national STI capacities. The present study outlines STI capacity around three sets of capabilities: technological capabilities, social capabilities, and common capabilities. The Global Science, Technology and Innovation Capacity (GSTIC) index was developed to provide current evidence on the national STI capacities of the countries, and to improve the composite indicators used for such purposes. The GSTIC ranks a large number of countries (167) on the basis of their STI capacities and categories them into four groups: i.e. leaders, dynamic adopters, slow adopters, and laggards. For more meaningful assessment of the STI capacities of nations, it captures the achievement gaps of individual countries with the highest achiever. The study also provides ranking and achievement gaps of nations in the nine GSTIC pillars: technology creation, R&D capacity, R&D performance, technology absorption, diffusion of old technologies, diffusion of recent innovations, exposure to foreign technology, human capital, and enabling factors. A more detailed analysis of the strengths and weaknesses in different pillars of STI capacity of ten selected countries is also provided. The results show that there are significant disparities among nations in STI capacity and its various aspects, and developing countries have much to catch-up with the developed nations. However, different countries may adopt different strategies according to their strengths and weaknesses. Useful insight into the strengths and weaknesses of the national STI capacities of different countries are provided in the study.

• 전자공학회논문지 IE
• /
• v.44 no.4
• /
• pp.1-10
• /
• 2007

In this thesis, we study the implementation of hands-free phone in a car, taking acoustic echo canceller, in order to remove acoustic echo effectively. Conventional coustic echo canceller used for only adaptive filtering has much difficulty to solve both echo and double-talk problem. To tackle this problem, we propose acoustic echo canceller consisting of adaptive filter using a modified NLMS, VAD to catch exact voice activity duration using two independent forgetting factors, double-talk detector to detect fast and precise double talk duration using cross-correlation between microphone signal and residual echo, and output controller using VAD and double-talk detector. The proposed hands-free phone taking acoustic echo canceller shows the performance that has not acoustic echo and guarantees full duplex.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.49 no.3
• /
• pp.367-375
• /
• 2016

Marbled flounder Pseudopleuronectes yokohamae is a commercially important flatfish in the coastal waters of the East Sea off Pohang, Korea. Maturity and spawning dynamics of P. yokohamae were investigated based on samples collected by gill net and long line in the coastal waters off Pohang from January 2013 to December 2014. We analyzed monthly changes in maturity stage, gonadosomatic index (GSI), egg diameter, fecundity, and total length at 50% group maturity. We found that the spawning period was December to February, wherein fecundity ranged from 55,760 eggs at 20.40 cm total length (TL) to 1,559,007 eggs at 46.6 cm TL. The relationship between TL and fecundity (F) was F = 0.0017 TL^3.5604 (R² = 0.6984), and F increased with increasing TL. We estimated the TL at 50% group maturity as 27.4 cm for female and 23.5 cm for male P. yokohamae, respectively. The results from this research will aid in the development of policies on season and minimum catch length for sustainable production of P. yokohamae. We propose that it is necessary to conduct long-term ecological monitoring for this species because biological properties such as spawning ecology and growth performance are being affected by climate change.

• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.1
• /
• pp.71-87
• /
• 2020

As startup companies are recognized as the important source of job creation and economic growth, the study is trying to find out the current situation of startup companies and their success factors in South Korea. The study analyzes the success factor of Woowa Brothers recognized as a unicorn start-up company in 2018 by using ERIS model and Start-up Developmental Stages. The ERIS model emphasizes Entrepreneurs, Resources, Industry Environment and Strategy for company performance and three developmental stages are divided into initiation, commercial and growth stages. The results indicate that the Woowa Brothers' success factor was to catch the appropriate timing in ERIS and to combine unique methods by differentiating itself from existing competitors at each stage. With analyses of success factor of Woowa Brothers as a case model, the study reveals limitations and recommends the direction of future research for producing more unicorn companies in South Korea.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.6
• /
• pp.741-747
• /
• 2010

These days many meeting minutes of various organizations are publicly available and the interest in these documents by people is increasing. However, it is time-consuming and tedious to read and understand whole documents even if the documents can be accessed easily. In addition, what most people want from meeting minutes is to catch the main issues of the meeting and understand its contexts rather than to know whole discussions of the meetings. This paper proposes a novel method for summarizing documents considering the characteristics of the meeting minutes. It first extracts the sentences which are addressing the main issues. For each issues expressed in the extracted sentences, the sentences related with the issue are then extracted in the next step. Then, by transforming the extracted sentences into a tree-structure form, the results of the proposed method can be understood better than existing methods. In the experiments, the proposed method shows remarkable improvement in performance and this result implies that the proposed method is plausible for summarizing meeting minutes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.9
• /
• pp.1081-1094
• /
• 2016

KREONET is a principal national R&E network running by KISTI in Korea. It uniquely provides production research network services for around 200 non-profit research and educational organizations, based on hybrid (IP and non-IP) network infrastructure. However, KREONET is limited to meet various needs of new network services for advanced Science & Technology (S&T) users because its infrastructure is inherently derived form classical hardware-based, fixed and closed environments. So, KREONET-S is designed to provide advanced S&T services to catch up with time-to-research and time-to-collaboration. In this paper, we present a system architecture of KREONET-S based on network infrastructure that consists of data and control planes separately. Furthermore, we propose and describe VDN service which is capable of building a virtual dedicate & bandwidth-guaranteed network for S&T group dynamically. we implement VDN application on KREONET-S and then perform performance analysis for proving that KREONET-S system and VDN application can be a good solutions to cope with new network paradigms for various advanced S&T applications and users.

• The KIPS Transactions:PartC
• /
• v.13C no.7 s.110
• /
• pp.843-850
• /
• 2006

This is devoted to first, to propose a hash chain algorithm that generates more secure key than conventional encryption method. Secondly, we proposes encryption method that is more secure than conventional system using a encryption method that encrypts each block with each key generated by a hash chain algorithm. Thirdly, After identifying the user via wired and wireless network using a user authentication method. We propose a divided session key method so that Although a client key is disclosed, Attackers cannot catch a complete key and method to safely transfer the key using a divided key method. We make an experiment using various size of digital contents files for performance analysis after performing the design and implementation of system. Proposed system can distribute key securely than conventional system and encrypt data to prevent attacker from decrypting complete data although key may be disclosed. The encryption and decryption time that client system takes to replay video data fie is analogous to the conventional method.

• Journal of Korea Technology Innovation Society
• /
• v.12 no.1
• /
• pp.237-264
• /
• 2009

Many Korean companies wanted to improve technological competitiveness and business performance radically through technology leadership initiatives. In-depth case studies about successful Korean technological innovation in the technology generation stage have potential to minimize Korea and developing country firms' trial and error when they are pursuing new technological innovation in the technology generation stage. There are few studies about developing country firms' technological innovations in the technology generation stage and especially process innovation studies are far less performed compared to product innovation studies. This is an exploratory study of POSCO's FINEX process technology innovation in the technology generation stage. These are my findings from this study. Firstly, leadership innovation in the technology generation stage is not a continuous development of catch-up innovation in the technology internalization stage and only top managements can initiate highly risky leadership innovation. Secondly, developing country firms which lacked in technological capability overcomes difficulties in the early stage through complementary technological collaboration with R&D first-movers. Thirdly, this company become a technology leader in spite of late entry in technology development race with developed country firms through rapid scale-ups.