• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.8
• /
• pp.359-368
• /
• 2017

What is the term "intelligence information society", which is a term that has been continuously discussed recently? This means that the automation beyond the limits of human ability in the whole societies based on intelligent information technology is a universalized social future. In particular, it is a concept that minimizes human intervention and continuously pursues evolution to data (or big data) -based automation. For example, autonomous automation is constantly aiming at unmanned vehicles with artificial intelligence as a key element. However, until now, intelligent information research has focused on the intelligence itself and has made an effort to improve intelligence logic and replace human brain and intelligence. On the other hand, in order to replace the human labor force, we have continued to make efforts to replace workers with robots by analyzing the working principles of workers and developing optimized simple logic. This study proposes important strategies and directions to implement intelligent information education policy and intelligent information technology research strategy by suggesting access strategy, education method and detailed policy road map for intelligent information technology research strategy and educational service. In particular, we propose a phased approach to intelligent information education such as basic intelligence education, intelligent content education, and intelligent application education. In addition, we propose education policy plan for the improvement of intelligent information technology, intelligent education contents, and intelligent education system as an important factor for success and failure of the 4th industrial revolution, which is centered on intelligence and automation.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.4
• /
• pp.355-360
• /
• 2015

In this paper, we propose the design of Radial Basis Function Neural Network based on PCA in order to recognize handwritten digits. The proposed pattern classifier consists of the preprocessing step of PCA and the pattern classification step of pRBFNNs. In the preprocessing step, Feature data is obtained through preprocessing step of PCA for minimizing the information loss of given data and then this data is used as input data to pRBFNNs. The hidden layer of the proposed classifier is built up by Fuzzy C-Means(FCM) clustering algorithm and the connection weights are defined as linear polynomial function. In the output layer, polynomial parameters are obtained by using Least Square Estimation (LSE). MNIST database known as one of the benchmark handwritten dataset is applied for the performance evaluation of the proposed classifier. The experimental results of the proposed system are compared with other existing classifiers.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.4
• /
• pp.1-13
• /
• 2023

This study develops an optimal mobile-communication-based National Highway traffic volume estimation model using an ensemble-based machine learning algorithm. Based on information such as mobile communication data and VDS data, the LightGBM model was selected as the optimal model for estimating traffic volume. As a result of evaluating traffic volume estimation performance from 96 points where VDS was installed, MAPE was 8.49 (accuracy 91.51%). On the roads where VDS was not installed, traffic estimation accuracy was 92.6%.

• Convergence Security Journal
• /
• v.23 no.3
• /
• pp.21-28
• /
• 2023

The development of information and communication technology in the 21st century has increased operational efficiency by providing hyper-connectivity and hyper-intelligence in the control systems of major infrastructure, but is also increasing security vulnerabilities, exposing it to hacking threats. Among them, the electric power system that supplies electric power essential for daily life has become a major target of cyber-attacks as a national critical infrastructure system. Recently, in order to protect these power systems, various security systems have been developed and the stability of the power systems has been maintained through practical cyber battle training. However, as cyber-attacks are combined with advanced ICT technologies such as artificial intelligence and big data, it is not easy to defend cyber-attacks that are becoming more intelligent with existing security systems. In order to defend against such intelligent cyber-attacks, it is necessary to know the types and aspects of intelligent cyber-attacks in advance. In this study, we analyzed the evolution of cyber attacks combined with advanced ICT technology.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.4
• /
• pp.189-198
• /
• 2023

As crimes frequently occur on the street, the spread of CCTV is increasing. However, due to the shortcomings of passively operated CCTV, the need for intelligent CCTV is attracting attention. Due to the heavy system of such intelligent CCTV, high-performance devices are required, which has a problem in that it is expensive to replace the general CCTV. To solve this problem, an intelligent CCTV system that recognizes low-quality images and operates even on devices with low performance is required. Therefore, this paper proposes a Saying CCTV system that can detect threats in real time by using the AWS cloud platform to lighten the system and convert images into text. Based on the data extracted using YOLO v4 and OpenPose, it is implemented to determine the risk object, threat behavior, and threat situation, and calculate the risk using machine learning. Through this, the system can be operated anytime and anywhere as long as the network is connected, and the system can be used even with devices with minimal performance for video shooting and image upload. Furthermore, it is possible to quickly prevent crime by automating meaningful statistics on crime by analyzing the video and using the data stored as text.

• Journal of Korea Society of Industrial Information Systems
• /
• v.17 no.4
• /
• pp.71-85
• /
• 2012

This paper aims to develop an event-driven failure analysis and prognosis system that is able to monitor ship status in real time, and efficiently react unforeseen system failures. In general, huge amount of recorded sensor data must be effectively interpreted for failure analysis, but unfortunately noise and redundant information in the gathered sensor data are obstacles to a successful analysis. This paper therefore applies 'Equal-frequency binning' and 'Entropy' techniques to extract only important information from the raw sensor data while minimizing information loss. The efficiency of the developed failure analysis system is demonstrated with the collected sensor data from a marine diesel engine.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.1
• /
• pp.55-63
• /
• 2023

In this paper, we propose a function and service of the Disaster Crisis Alert Management System that automatically analyzes the situation judgment criteria to issue a disaster crisis alert and a plan to operate in the National Disaster Management System(NDMS). In the event of a disaster, a crisis alert(interest-caution-alert-serious) is issued according to the crisis alert level. In order to automatically analyze and determine the crisis alert level, first, data collection, crisis alert level analysis, crisis alert level judgment, and disaster crisis alert management system that expresses the crisis alert level by spatial scale(province, city, district) were implemented. The crisis alert level was analyzed and expressed in two ways by applying the intelligent crisis alert level(determination of regional sensitivity, risk level, and crisis alert level) and the crisis alert standard of the crisis management manual(province-level standard setting). Second, standard metadata, linkage of situation information of target) and API standards for data provision are presented to jointly utilize data linkage and crisis alert data of the disaster and safety data sharing platform so that it can be operated within the NDMS.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.97-113
• /
• 2016

Due to the growth of internet data and the rapid development of internet technology, "big data" analysis is actively conducted to analyze enormous data for various purposes. Especially in recent years, a number of studies have been performed on the applications of text mining techniques in order to overcome the limitations of existing structured data analysis. Various studies on sentiment analysis, the part of text mining techniques, are actively studied to score opinions based on the distribution of polarity of words in documents. Usually, the sentiment analysis uses sentiment dictionary contains positivity and negativity of vocabularies. As a part of such studies, this study tries to construct sentiment dictionary which is customized to specific data domain. Using a common sentiment dictionary for sentiment analysis without considering data domain characteristic cannot reflect contextual expression only used in the specific data domain. So, we can expect using a modified sentiment dictionary customized to data domain can lead the improvement of sentiment analysis efficiency. Therefore, this study aims to suggest a way to construct customized dictionary to reflect characteristics of data domain. Especially, in this study, movie review data are divided by genre and construct genre-customized dictionaries. The performance of customized dictionary in sentiment analysis is compared with a common sentiment dictionary. In this study, IMDb data are chosen as the subject of analysis, and movie reviews are categorized by genre. Six genres in IMDb, 'action', 'animation', 'comedy', 'drama', 'horror', and 'sci-fi' are selected. Five highest ranking movies and five lowest ranking movies per genre are selected as training data set and two years' movie data from 2012 September 2012 to June 2014 are collected as test data set. Using SO-PMI (Semantic Orientation from Point-wise Mutual Information) technique, we build customized sentiment dictionary per genre and compare prediction accuracy on review rating. As a result of the analysis, the prediction using customized dictionaries improves prediction accuracy. The performance improvement is 2.82% in overall and is statistical significant. Especially, the customized dictionary on 'sci-fi' leads the highest accuracy improvement among six genres. Even though this study shows the usefulness of customized dictionaries in sentiment analysis, further studies are required to generalize the results. In this study, we only consider adjectives as additional terms in customized sentiment dictionary. Other part of text such as verb and adverb can be considered to improve sentiment analysis performance. Also, we need to apply customized sentiment dictionary to other domain such as product reviews.

• Journal of Broadcast Engineering
• /
• v.26 no.5
• /
• pp.533-542
• /
• 2021

In this study, a process of re-creating Jindo Buk Chum, one of the traditional Korean arts, into digital art using various artificial intelligence technologies was proposed. The audience's emotional data, quantified through artificial intelligence language analysis technology, intervenes in various object forms in the projection mapping performance and affects the big story without changing it. If most interactive arts express communication between the performer and the video, this performance becomes a new type of responsive performance that allows the audience to directly communicate with the work, centering on artificial intelligence emotion analysis technology. This starts with 'Chuimsae', a performance that is common only in Korean traditional art, where the audience directly or indirectly intervenes and influences the performance. Based on the emotional information contained in the performer's 'prologue', it is combined with the audience's emotional information and converted into the form of images and particles used in the performance to indirectly participate and change the performance.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.2
• /
• pp.191-196
• /
• 2024

Recently, research on intelligent IoT technology has been actively conducted by various companies and research institutes to analyze various data collected from IoT devices and provide it through actual application services. However, security issues such as personal information leakage may arise in the process of transmitting and receiving data to use data collected from IoT devices for research and development. In addition, as data collected from multiple IoT devices increases, data management difficulties exist, and data movement is costly and time consuming. Therefore, in this paper, we intend to develop an Ethereum network-based federated learning system with guaranteed reliability to improve security issues and inefficiencies in a federated learning environment composed of various devices.