• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.2
• /
• pp.742-762
• /
• 2015

In this study, we propose a distance metric learning approach called discriminant metric learning (DML) for face verification, which addresses a binary-class problem for classifying whether or not two input images are of the same subject. The critical issue for solving this problem is determining the method to be used for measuring the distance between two images. Among various methods, the large margin nearest neighbor (LMNN) method is a state-of-the-art algorithm. However, to compensate the LMNN's entangled data distribution due to high levels of appearance variations in unconstrained environments, DML's goal is to penalize violations of the negative pair distance relationship, i.e., the images with different labels, while being integrated with LMNN to model the distance relation between positive pairs, i.e., the images with the same label. The likelihoods of the input images, estimated using DML and LMNN metrics, are then weighted and combined for further analysis. Additionally, rather than using the k-nearest neighbor (k-NN) classification mechanism, we propose a verification mechanism that measures the correlation of the class label distribution of neighbors to reduce the false negative rate of positive pairs. From the experimental results, we see that DML can modify the relation of negative pairs in the original LMNN space and compensate for LMNN's performance on faces with large variances, such as pose and expression.

• Journal of Information Processing Systems
• /
• v.18 no.2
• /
• pp.245-257
• /
• 2022

In the Internet of Things (IoT) era, the types of devices used by one user are becoming more diverse and the number of devices is also increasing. However, a forensic investigator is restricted to exploit or collect all the user's devices; there are legal issues (e.g., privacy, jurisdiction) and technical issues (e.g., computing resources, the increase in storage capacity). Therefore, in the digital forensics field, it has been a challenge to acquire information that remains on the devices that could not be collected, by analyzing the seized devices. In this study, we focus on the fact that multiple devices share data through account synchronization of the online platform. We propose a novel way of identifying the user's interest through analyzing the remnants of targeted advertising which is provided based on the visited websites or search terms of logged-in users. We introduce a detailed methodology to pick out the targeted advertising from cache data and infer the user's interest using deep learning. In this process, an improved learning model considering the unique characteristics of advertisement is implemented. The experimental result demonstrates that the proposed method can effectively identify the user interest even though only one device is examined.

• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.32-44
• /
• 2023

Information security breach is a major risk in e-learning. This study presents the potential information security disruptions in Learning Management Systems (LMS) from the perspective of users. We use the Technology Acceptance Model approach as a user perception model of information security, and the results of a questionnaire comprising 44 questions for instructors and students across Indonesia to verify the model. The results of the data analysis and model testing reveals that lecturers and students perceive the level of information security in the LMS differently. In general, the information security aspects of LMSs affect the perceptions of trust of student users, whereas such a correlation is not found among lecturers. In addition, lecturers perceive information security aspect on Moodle is and Google Classroom differently. Based on this finding, we recommend that institutions make more intense efforts to increase awareness of information security and to run different information security programs.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.77-91
• /
• 2022

Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, such as system identification and damage detection. Therefore, developing efficient techniques is quite essential to recognize the anomalies in monitoring data. In this study, several machine learning techniques are explored and implemented to detect and classify various types of data anomalies. A field dataset, which consists of one month long acceleration data obtained from a long-span cable-stayed bridge in China, is employed to examine the machine learning techniques for automated data anomaly detection. These techniques include the statistic-based pattern recognition network, spectrogram-based convolutional neural network, image-based time history convolutional neural network, image-based time-frequency hybrid convolution neural network (GoogLeNet), and proposed ensemble neural network model. The ensemble model deliberately combines different machine learning models to enhance anomaly classification performance. The results show that all these techniques can successfully detect and classify six types of data anomalies (i.e., missing, minor, outlier, square, trend, drift). Moreover, both image-based time history convolutional neural network and GoogLeNet are further investigated for the capability of autonomous online anomaly classification and found to effectively classify anomalies with decent performance. As seen in comparison with accuracy, the proposed ensemble neural network model outperforms the other three machine learning techniques. This study also evaluates the proposed ensemble neural network model to a blind test dataset. As found in the results, this ensemble model is effective for data anomaly detection and applicable for the signal characteristics changing over time.

• Journal of Korea Society of Industrial Information Systems
• /
• v.15 no.5
• /
• pp.235-246
• /
• 2010

Even after Hwang Woo Suk's research ethics scandal in Korean research fields, there are still research ethic problems that are present in the Korean society. In order to spread research ethics education around Korea, there needs to be sufficient amount of research ethics contents prepared, which then can be used to educate and put into action. Through the usages of the scope of research ethics, comparing and contrasting several examples from domestic illegal research acts and American research ethics education, and the Learning Activity Management Systems (LAMS), we would be able to develop a way to manage the research ethics education. Also, in order to fully use and apply the necessary contents and material for the research ethics education program in several universities, we would not only need to use the LAMS system to establish the training system, but also to spread the awareness of research ethics.

• Korean Medical Education Review
• /
• v.16 no.1
• /
• pp.16-24
• /
• 2014

The development of information technology (IT) has led to changes in medical education. IT has been used for e-learning and e-teaching, e-assessment, e-logistics, and e-administration. Pusan National University School of Medicine has developed its own educational information system using IT to support students' learning and assessment and to manage curricular activities. Based on our experience, we propose six suggestions for designing new software, specifically regarding simplifying the design for users, communication with the programmer, a rapid cycle from design to implementation, orientation support for users, backup and security support, and obtaining patents. Collaborating with the Department of Medical Informatics within the School of Medicine has advantages, and an alliance among medical schools can simplify the development of software. In any case, curricular innovation should precede the introduction of technology because all technologies should support curricular goals.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.2
• /
• pp.227-234
• /
• 2012

Development of Computer science and internet technoloy have led changes all over the social area. Educational markets based on this circumstance are offering various services named remote education, cyber lecture, e-Learning, etc. Due to these products, systems for computer based teaching and evaluating student's achievement are wide spread. But in many systems we can find functional restrictions. In this paper we propose a RTE system offering interactive self control mode based education so as to provide customized education for each individual by realtime feedback of the level of the student's comprehension we expect that this system provides customized education environment considering student's achievement level and maximizes their motivation.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.406-411
• /
• 2005

In this paper, we present the area-based Q-learning to search a target object using multiple robot. To search the target in Markovian space, the robots should recognize their surrounding at where they are located and generate some rules to act upon by themselves. Under area-based Q-learning, a robot, first of all, obtains 6-distances from itself to environment by infrared sensor which are hexagonally allocated around itself. Second, it calculates 6-areas with those distances then take an action, i.e., turn and move toward where the widest space will be guaranteed. After the action is taken, the value of Q will be updated by relative formula at the state. We set up an experimental environment with five small mobile robots, obstacles, and a target object, and tried to search for a target object while navigating in a unknown hallway where some obstacles were placed. In the end of this paper, we presents the results of three algorithms - a random search, area-based action making (ABAM), and hexagonal area-based Q-teaming.

• Phonetics and Speech Sciences
• /
• v.12 no.2
• /
• pp.29-37
• /
• 2020

Recently, the neural network-based deep learning algorithm has dramatically improved performance compared to the classical Gaussian mixture model based hidden Markov model (GMM-HMM) automatic speech recognition (ASR) system. In addition, researches on end-to-end (E2E) speech recognition systems integrating language modeling and decoding processes have been actively conducted to better utilize the advantages of deep learning techniques. In general, E2E ASR systems consist of multiple layers of encoder-decoder structure with attention. Therefore, E2E ASR systems require data with a large amount of speech-text paired data in order to achieve good performance. Obtaining speech-text paired data requires a lot of human labor and time, and is a high barrier to building E2E ASR system. Therefore, there are previous studies that improve the performance of E2E ASR system using relatively small amount of speech-text paired data, but most studies have been conducted by using only speech-only data or text-only data. In this study, we proposed a semi-supervised training method that enables E2E ASR system to perform well in corpus in different domains by using both speech or text only data. The proposed method works effectively by adapting to different domains, showing good performance in the target domain and not degrading much in the source domain.

• Journal of Information Processing Systems
• /
• v.6 no.4
• /
• pp.609-620
• /
• 2010

In order to provide a tailored education for learners within the ubiquitous environment, it is critical to undertake an analysis of the learning activities of learners. For this purpose, SCORM (Sharable Contents Object Reference Model), IMS LD (Instructional Management System Learning Design) and other standards provide learning design support functions, such as, progress checks. However, in order to apply these types of standards, contents packaging is required, and due to the complicated standard dimensions, the facilitation level is lower than the work volume when developing the contents and this requires additional work when revision becomes necessary. In addition, since the learning results are managed by the server there is the problem of the OS being unable to save data when the network is cut off. In this study, a system is realized to manage the actions of learners through the event interception of a web-browser by using event hooking. Through this technique, all HTMLbased contents can be facilitated again without additional work and saving and analysis of learning results are available to improve the problems following the application of standards. Furthermore, the ubiquitous learning environment can be supported by tracking down learning results when the network is cut off.