• Journal of Intelligence and Information Systems
• /
• v.16 no.1
• /
• pp.93-116
• /
• 2010

Ensemble approach is applied to the detection modeling of illegal cash accommodation (ICA) that is the well-known type of fraudulent usages of credit cards in far east nations and has not been addressed in the academic literatures. The performance of fraud detection model (FDM) suffers from the imbalanced data problem, which can be remedied to some extent using an ensemble of many classifiers. It is generally accepted that ensembles of classifiers produce better accuracy than a single classifier provided there is diversity in the ensemble. Furthermore, recent researches reveal that it may be better to ensemble some selected classifiers instead of all of the classifiers at hand. For the effective detection of ICA, we adopt ensemble size reduction technique that prunes the ensemble of all classifiers using accuracy and diversity measures. The diversity in ensemble manifests itself as disagreement or ambiguity among members. Data imbalance intrinsic to FDM affects our approach for ICA detection in two ways. First, we suggest the training procedure with over-sampling methods to obtain diverse training data sets. Second, we use some variants of accuracy and diversity measures that focus on fraud class. We also dynamically calculate the diversity measure-Forward Addition and Backward Elimination. In our experiments, Neural Networks, Decision Trees and Logit Regressions are the base models as the ensemble members and the performance of homogeneous ensembles are compared with that of heterogeneous ensembles. The experimental results show that the reduced size ensemble is as accurate on average over the data-sets tested as the non-pruned version, which provides benefits in terms of its application efficiency and reduced complexity of the ensemble.

• Journal of The Korean Association of Information Education
• /
• v.25 no.6
• /
• pp.973-985
• /
• 2021

The purpose of this study is to suggest the direction and implications of learning with AI in the future by analyzing the trends of research learning with AI in the field of education. For doing this, the final 78 papers published in domestic journals over the past three years from 2019 to July 2021 were selected for analysis through review. The analysis results are as follows. First of all, papers in 2020 among the three years were most published, and the most utilized research method was the qualitative research. In addition, according to the analysis by study subject, studies on elementary school students were the most common, followed by studies on college and graduate students. In the analysis by subject, research related to foreign language education was most utilized and chatbot was most used in the AI technology type. Finally, the research learning with AI accounted for the majority, and student support accounted for the majority as the type of education system learning with AI at the implementation stage among the areas of teaching and learning and evaluation. Based on these results, the direction and implications of learning with AI in the future were presented. This study is meaningful in that it grasped research trends of learning with AI in domestic from an overall perspective, and examined learning with AI focusing on the instructor-learner and the teaching and learning design process.

• KIISE Transactions on Computing Practices
• /
• v.23 no.12
• /
• pp.651-659
• /
• 2017

IT(Information Technology) has focused on infrastructure-technologies in the past but now focuses on IT service. Many companies strive to save costs and improve IT services. For this reason, they strive to implement a good internalization of ITSM (IT Service Management) by developing ITSM systems based on ITIL (Information Technology Infrastructure Library) from the late 2000s. In particular, IT service operations are one of the structure elements of ITIL version 3 and are highly related to the internalization of ITSM. However, in spite-of-the successful implementation of ITSM, the efficiency of IT service management has not improved due to iterative issues. Therefore, this study developed a user-centered training system by defining the iterative issue guidelines and implementing a database. The implemented user-centered training system provided IT service users with regular training services to produce a good solution for iterative issues after connecting the operation part of the ITSM system. Based on the results of this study, we expect that the proposed ITSM system will contribute to efficiently managing IT services by improving the limitations of IT service operation in the ITSM system.

• KIISE Transactions on Computing Practices
• /
• v.23 no.1
• /
• pp.57-64
• /
• 2017

Short message service (SMS) is one of the most important communication methods for people who use mobile phones. However, illegal advertising spam messages exploit people because they can be used without the need for friend registration. Recently, spam message filtering systems that use machine learning have been developed, but they have some disadvantages such as requiring many calculations. In this paper, we implemented a spam message filtering system using the set-based POI search algorithm and sentence similarity without servers. This algorithm can judge whether the input query is a spam message or not using only letter composition without any server computing. Therefore, we can filter the spam message although the input text message has been intentionally modified. We added a specific preprocessing option which aims to enable spam filtering. Based on the experimental results, we observe that our spam message filtering system shows better performance than the original set-based POI search algorithm. We evaluate the proposed system through extensive simulation. According to the simulation results, the proposed system can filter the text message and show high accuracy performance against the text message which cannot be filtered by the 3 major telecom companies.

• Journal of Cadastre & Land InformatiX
• /
• v.53 no.2
• /
• pp.53-66
• /
• 2023

Recently, there has been a significant interest in the development of autonomous driving simulation environment based on digital twin. In the development of such digital twin-based simulation environment, many researches has been conducted not only performance and functionality validation of autonomous driving, but also generation of virtual training data for deep learning. However, such digital twin-based autonomous driving simulation system has the problem of requiring a significant amount of time and cost for the system development and the data construction. Therefore, in this research, we aim to propose a method for rapidly designing and implementing a digital twin-based autonomous driving simulation system, using only the existing 3D models and high-definition map. Specifically, we propose a method for integrating 3D model of FBX and NGII HD Map for the Busan EDC area into CARLA, and a method for adding and modifying CARLA functions. The results of this research show that it is possible to rapidly design and implement the simulation system at a low cost by using the existing 3D models and NGII HD map. Also, the results show that our system can support various functions such as simulation scenario configuration, user-defined driving, and real-time simulation of traffic light states. We expect that usability of the system will be significantly improved when it is applied to broader geographical area in the future.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.5
• /
• pp.53-63
• /
• 2015

Singing voice synthesis is the generation of a song using a computer given its lyrics and musical notes. Hidden Markov models (HMM) have been proved to be the models of choice for text to speech synthesis. HMMs have also been used for singing voice synthesis research, however, a huge database is needed for the training of HMMs for singing voice synthesis. And commercially available singing voice synthesis systems which use the piano roll music notation, needs to adopt the easy to read standard music notation which make it suitable for singing learning applications. To overcome this problem, we use a speech database for training context dependent HMMs, to be used for singing voice synthesis. Pitch and duration control methods have been devised to modify the parameters of the HMMs trained on speech, to be used as the synthesis units for the singing voice. This work describes a singing voice synthesis system which uses a MusicXML based music score editor as the front-end interface for entry of the notes and lyrics to be synthesized and a hidden Markov model based text to speech synthesis system as the back-end synthesizer. A perceptual test shows the feasibility of our proposed system.

• Journal of Digital Convergence
• /
• v.20 no.2
• /
• pp.289-297
• /
• 2022

The purpose of this study was to investigate the perception of dental students based on their experiences of online classes after taking non-face-to-face education courses for all the school semesters in 2020. For the research method, an online survey was conducted on A survey was conducted on 161 dental students enrolled in A University. The analytical method was conducted through frequency analysis, correlation analysis, and multiple regression analysis. The survey analysis findings showed that the satisfaction of dental students' about the non-face-to-face education course was above 4.2, and the detailed items were in the order of the appropriateness of the attendance processing method, satisfaction with recorded video lectures, and the assessment method of the course grade. In the case of the factors that affect the satisfaction of non-face-to-face education courses, the learning system and assessment method were statistically significant. The online class type that is most preferred by the students is recorded video lectures, and the highest number of participants chose 21~30 minutes as the appropriate time for the class content. It is considered that the application of the online system will continue to be used together with face-to-face education courses in the education site and various university-level efforts like systematic support are required to achieve effective learning achievements. This study only investigated the non-face-to-face education operation conditions of A University, so it cannot be generalized to all universities, but it can be used as basic data to provide education curriculum design and supportive measures for the compatibility of face-to-face and non-face-to-face courses.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.276-279
• /
• 2022

The 4th industrial revolution refers to the transition to a knowledge society in which the production speed of knowledge is fast and the proportion of the knowledge industry is greatly increased. The reorganization of the industrial structure and the change of occupations and jobs due to new technologies are bringing about changes in education, and the development of digital technology has made education that is borderless, individual and dynamic, becoming the new standard of education. With these changes, interest in nano-degrees on new technologies or micro-degrees focused on core courses rather than regular course degrees is increasing. As a representative example, Udacity in the United States has opened and operated online nanodegree courses related to jobs, and collaborates with major companies to develop and educate core training courses necessary for companies, thereby efficiently supporting companies in securing talent. With the revitalization of online vocational and job training, an environment has been prepared in which individuals can set goals for vocational competency development and continue their portfolio-based sustainable learning. However, for effective vocational education, automated and personalized educational content design should be preceded. To this end, in this paper, we propose a personalized career and course map recommendation system in the era of online learning

• Journal of the Economic Geographical Society of Korea
• /
• v.17 no.1
• /
• pp.45-68
• /
• 2014

This research addressed improvement directions and problems of regional science and technology(S&T) policies in Daegu and Gyungbuk regions in terms of 'Creative Economy' which was a new paradigm of Park Gunhye Government. Creative Economy stressed the construction of an ecosystem in regional research and development activities, and thus it was deeply associated with building a regional innovation system(RIS). There were several problems to strengthen RIS with regional S&T policies of the regions as follows: limits in meeting regional needs due to excessive attraction of central government's projects into regions; the high ratio of programs for future basic research potentials; the lack of programs assessing and coordinating the policies; and the lack of experiences and expenditure of research institutes and firm supporting organizations. Due to these problems, the role of the policies in building RIS did not seem to be effective. Therefore, the policies need to be improved through the following measures: the expansion of regional own policies focusing regional needs; the enhancement of policy coordination by shifting to systematic approach; the expansion of soft supporting programs for constructing innovation systems; and the enhancement of stability and ability of research institutes and firm supporting organizations.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.1
• /
• pp.58-68
• /
• 2010

In this paper, we propose an efficient link protection switching scheme for provider backbone bridge systems with a spanning tree for backup links exclusively, and evaluate its performance. The proposed scheme offers guaranteed QoS flows even when a link fault occurrs in the primary link by flooding the flows over the profiled spanning tree. The flooding mechanism over the spanning tree can also provide low latency and remove the loopback flows. We also derive the efficiency of bandwidth usage for the normal flows and the number of lost frames during the link restoration. For evaluating its feasibility, we implement a prototype of PBB-TE systems based on the Linux bridge codes, which can support both link protection switching capability with CCM and MAC-in-MAC encapsulation. A related protocol analyzer is also developed. One can see that the proposed scheme and the prototype can be useful for developing carrier class Ethernet systems based on PBB-TE.