• KIISE Transactions on Computing Practices
• /
• v.23 no.8
• /
• pp.466-472
• /
• 2017

Hardware Transactional Memory (HTM) has greatly changed the parallel programming paradigm for transaction processing. Since Intel has recently proposed Transactional Synchronization Extension (TSX), a number of studies based on HTM have been conducted. However, the existing studies support conflict prediction for a single cause of the transaction processing and provide a standardized TSX environment for all workloads. To solve the problems, we propose an efficient hardware transactional memory scheme for processing transactions in multi-core in-memory environment. First, the proposed scheme determines whether to use Software Transactional Memory (STM) or the serial execution as a fallback path of HTM by using a prediction matrix to collect the information of previously executed transactions. Second, the proposed scheme performs efficient transaction processing according to the characteristic of a given workload by providing a retry policy based on machine learning algorithms. Finally, through the experimental performance evaluation using Stanford transactional applications for multi-processing (STAMP), the proposed scheme shows 10~20% better performance than the existing schemes.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.3
• /
• pp.611-618
• /
• 2015

By Wikipedia, big data is a broad term for data sets so large or complex that traditional data processing applications are inadequate. Data mining is the computational process of discovering patterns in huge data sets involving methods at the intersection of association rule, decision tree, clustering, artificial intelligence, machine learning. Association rule is a well researched method for discovering interesting relationships between itemsets in huge databases and has been applied in various fields. There are positive, negative, and inverse association rules according to the direction of association. If you want to set the evaluation criteria of association rule, it may be desirable to consider three types of association rules at the same time. To this end, we proposed a balanced comparative confidence considering sensitivity, specificity, false positive, and false negative, checked the conditions for association threshold by Piatetsky-Shapiro, and compared it with comparative confidence and inversely comparative confidence through a few experiments.

• Smart Structures and Systems
• /
• v.26 no.1
• /
• pp.103-115
• /
• 2020

The cold-formed steel storage racks are extensively employed in various industries applications such as storing products in reliable places and storehouses before distribution to the market. Racking systems lose their stability under lateral loads, such as seismic actions due to the slenderness of elements and low ductility. This justifies a need for more investigation on methods to improve their behavior and increase their capacity to survive medium to severe loads. A standardized connection could be obtained through investigation on the moment resistance, value of original rotational stiffness, ductility, and failure mode of the connection. A total of six monotonic tests were carried out to determine the behavior of the connection of straight 2.0 mm, and 2.6 mm thickness connects to 5 lug end connectors. Then, the obtained results are benched mark as the original data. Furthermore, an extreme learning machine (ELM) technique has been employed to verify and predict both moment and rotation results. Out of 4 connections, increase the ultimate moment resistance of connection by 13% and 18% for 2.0 mm and 2.6 mm upright connection, respectively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.3
• /
• pp.1168-1188
• /
• 2018

Social Network Aggregators are used to maintain and manage manifold accounts over multiple online social networks. Displaying the Activity feed for each social network on a common dashboard has been the status quo of social aggregators for long, however retrieving the desired data from various social networks is a major concern. A user inputs the query desiring the specific outcome from the social networks. Since the intention of the query is solely known by user, therefore the output of the query may not be as per user's expectation unless the system considers 'user-centric' factors. Moreover, the quality of solution depends on these user-centric factors, the user inclination and the nature of the network as well. Thus, there is a need for a system that understands the user's intent serving structured objects. Further, choosing the best execution and optimal ranking functions is also a high priority concern. The current work finds motivation from the above requirements and thus proposes the design of a query processing system to retrieve information from social network that extracts user's intent from various social networks. For further improvements in the research the machine learning techniques are incorporated such as Latent Dirichlet Algorithm (LDA) and Ranking Algorithm to improve the query results and fetch the information using data mining techniques.The proposed framework uniquely contributes a user-centric query retrieval model based on natural language and it is worth mentioning that the proposed framework is efficient when compared on temporal metrics. The proposed Query Processing System to Retrieve Information from Social Network (QPSSN) will increase the discoverability of the user, helps the businesses to collaboratively execute promotions, determine new networks and people. It is an innovative approach to investigate the new aspects of social network. The proposed model offers a significant breakthrough scoring up to precision and recall respectively.

• Journal of the Korea Computer Graphics Society
• /
• v.24 no.1
• /
• pp.19-26
• /
• 2018

In this paper, we propose a gaze depth estimation method based on a binocular eye tracker for virtual reality and augmented reality applications. The proposed gaze depth estimation method collects a wide range information of each eye from the eye tracker such as the pupil center, gaze direction, inter pupil distance. It then builds gaze estimation models using Multilayer perceptron which infers gaze depth with respect to the eye tracking information. Finally, we evaluated the gaze depth estimation method with 13 participants in two ways: the performance based on their individual models and the performance based on the generalized model. Through the evaluation, we found that the proposed estimation method recognized gaze depth with 90.1% accuracy for 13 individual participants and with 89.7% accuracy for including all participants.

• Journal of KIISE:Software and Applications
• /
• v.34 no.10
• /
• pp.889-899
• /
• 2007

In proteomics, recent advancements In mass spectrometry technology and in protein extraction and separation technology made high-throughput analysis possible. This leads to thousands to hundreds of thousands of MS/MS spectra per single LC-MS/MS experiment. Such a large amount of data creates significant computational challenges and therefore effective data analysis methods that make efficient use of computational resources and, at the same time, provide more peptide identifications are in great need. Here, SIFTER system is designed to avoid inefficient processing of shotgun proteomic data. SIFTER provides software tools that can improve throughput of mass spectrometry-based peptide identification by filtering out poor-quality tandem mass spectra and estimating a Peptide charge state prior to applying analysis algorithms. SIFTER tools characterize and assess spectral features and thus significantly reduce the computation time and false positive rates by localizing spectra that lead to wrong identification prior to full-blown analysis. SIFTER enables fast and in-depth interpretation of tandem mass spectra.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.9
• /
• pp.615-620
• /
• 2013

With the development of smartphones, the number of applications for smartphone increases sharply. As a result, users need to try several times to find their favorite apps. In order to solve this problem, we propose a recommendation system to provide an appropriate app list based on the user's log information including time stamp, location, application list, and so on. The proposed approach learns three recommendation models including Naive-Bayesian model, SVM model, and Most-Frequent Usage model using temporal and spatial attributes. In order to figure out the best model, we compared the performance of these models with variant features, and suggest an hybrid method to improve the performance of single models.

• Journal of KIISE:Software and Applications
• /
• v.32 no.12
• /
• pp.1260-1270
• /
• 2005

Support Vector Machines (SVM) is well known for a representative learner as one of the kernel methods. SVM which is based on the statistical learning theory shows good generalization performance and has been applied to various pattern recognition problems. However, SVM is basically to deal with a two-class classification problem, so we cannot solve directly a multi-class problem with a binary SVM. One-Per-Class (OPC) and All-Pairs have been applied to solve the face recognition problem, which is one of the multi-class problems, with SVM. The two methods above are ones of the output coding methods, a general approach for solving multi-class problem with multiple binary classifiers, which decomposes a complex multi-class problem into a set of binary problems and then reconstructs the outputs of binary classifiers for each binary problem. In this paper, we introduce the output coding methods as an approach for extending binary SVM to multi-class SVM and propose new output coding schemes based on the Error-Correcting Output Codes (ECOC) which is a dominant theoretical foundation of the output coding methods. From the experiment on the face recognition, we give empirical results on the properties of output coding methods including our proposed ones.

• The Journal of Korean Association of Computer Education
• /
• v.23 no.1
• /
• pp.37-44
• /
• 2020

As the importance of artificial intelligence(AI) education is emphasized recently, policies and researches are being promoted to develop the AI curriculum or courses for K-12 students in worldwide. In this study, researcher analysed a synthesis of contents and standards on AI education curriculum to present implications for AI education in the elementary and secondary schools. As a result, Korea and the United States are proposing national curriculum standards to provide the basis for AI curriculum establishment in school sites and to provide guidelines for various related policies such as teacher training programs. The EU's AI education is characterized by its curriculum and online courses to ensure that all citizens of the EU have AI literacy, rather than designating students or subjects at specific school levels. In terms of educational contents and levels, Korea, United States, and EU's curriculum or standards includes basics and applications related to machine learning and neural network based on the fundamental concepts and principles of artificial intelligence.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.4
• /
• pp.292-302
• /
• 2020

Machine learning algorithms have made immense contributions in various fields including sonar and radar applications. Recently developed Cycle-Consistency Generative Adversarial Network (CycleGAN), a variant of GAN has been successfully used for unpaired image-to-image translation. We present a modified CycleGAN for translation of underwater ship engine sounds with high perceptual quality. The proposed network is composed of an improved generator model trained to translate underwater audio from one vessel type to other, an improved discriminator to identify the data as real or fake and a modified cycle-consistency loss function. The quantitative and qualitative analysis of the proposed CycleGAN are performed on publicly available underwater dataset ShipsEar by evaluating and comparing Mel-cepstral distortion, pitch contour matching, nearest neighbor comparison and mean opinion score with existing algorithms. The analysis results of the proposed network demonstrate the effectiveness of the proposed network.