• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.10
• /
• pp.2643-2657
• /
• 2023

Scene graphs are structured representations that can clearly convey objects and the relationships between them, but are often heavily biased due to the highly skewed, long-tailed relational labeling in the dataset. Indeed, the visual world itself and its descriptions are biased. Therefore, Unbiased Scene Graph Generation (USGG) prefers to train models to eliminate long-tail effects as much as possible, rather than altering the dataset directly. To this end, we propose Geometric and Semantic Improvement (GSI) for USGG to mitigate this issue. First, to fully exploit the feature information in the images, geometric dimension and semantic dimension enhancement modules are designed. The geometric module is designed from the perspective that the position information between neighboring object pairs will affect each other, which can improve the recall rate of the overall relationship in the dataset. The semantic module further processes the embedded word vector, which can enhance the acquisition of semantic information. Then, to improve the recall rate of the tail data, the Class Balanced Seesaw Loss (CBSLoss) is designed for the tail data. The recall rate of the prediction is improved by penalizing the body or tail relations that are judged incorrectly in the dataset. The experimental findings demonstrate that the GSI method performs better than mainstream models in terms of the mean Recall@K (mR@K) metric in three tasks. The long-tailed imbalance in the Visual Genome 150 (VG150) dataset is addressed better using the GSI method than by most of the existing methods.

• Korea Trade Review
• /
• v.47 no.6
• /
• pp.335-353
• /
• 2022

There are two aspects of digital trade: the digitalisation of goods/services being traded and the digitalisation of the transactional act. Digital data (i.e. machine-readable industrial data and transactional data) is the major driving force for both aspects of digital trade. Digital data is a non-rivalrous input, whether for production or marketing activities, and is thus able to be used by many firms or government agencies without limiting the use of others. Digital platforms provide online infrastructure for the interactions between groups, for instance, consumers and producers. The externality effect refers to the situation in which prosperity in one group on a given platform will improve the returns of other groups on the same platform. In the era of the data-driven economy, strategic trade policy can involve data-related policies. The major objective of these policies is to improve the competitiveness of domestic firms. For instance, firms may be subsidised if they use cloud services provided by specific platforms. This strand of strategic trade policies might be useful for increasing the competitiveness of small-and medium-sized enterprises (SMEs) via the digitalisation of production/marketing processes. Alternatively, strategic trade policy may also exploit the externality effect via platform economy-related policies. Further, some countries may form data coalitions to facilitate cross-border data flow. This paper uses cases in Asian countries to illustrate which role these strategic trade policies can play in the digital economy.

• Journal of Internet of Things and Convergence
• /
• v.9 no.6
• /
• pp.1-9
• /
• 2023

Malware files containing concealed malicious scripts have recently been identified within MS Office documents frequently. In response, this paper describes the design and implementation of a system that automatically detects malicious digital files using machine learning techniques. The system is proficient in identifying malicious scripts within MS Office files that exploit the OLE VBA macro functionality, detecting malicious scripts embedded within the CDH/LFH/ECDR internal field values through OOXML structure analysis, and recognizing abnormal CDH/LFH information introduced within the OOXML structure, which is not conventionally referenced. Furthermore, this paper presents a mechanism for utilizing the VirusTotal malicious script detection feature to autonomously determine instances of malicious tampering within MS Office files. This leads to the design and implementation of a machine learning-based integrated software. Experimental results confirm the software's capacity to autonomously assess MS Office file's integrity and provide enhanced detection performance for arbitrary MS Office files when employing the optimal machine learning model.

• Convergence Security Journal
• /
• v.23 no.5
• /
• pp.65-72
• /
• 2023

This paper proposes a method for detecting malicious domains considering human habitual characteristics by building a Deep Learning model based on LSTM (Long Short-Term Memory). DGA (Domain Generation Algorithm) malicious domains exploit human habitual errors, resulting in severe security threats. The objective is to swiftly and accurately respond to changes in malicious domains and their evasion techniques through typosquatting to minimize security threats. The LSTM-based Deep Learning model automatically analyzes and categorizes generated domains as malicious or benign based on malware-specific features. As a result of evaluating the model's performance based on ROC curve and AUC accuracy, it demonstrated 99.21% superior detection accuracy. Not only can this model detect malicious domains in real-time, but it also holds potential applications across various cyber security domains. This paper proposes and explores a novel approach aimed at safeguarding users and fostering a secure cyber environment against cyber attacks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.1042-1058
• /
• 2024

Terahertz (THz) communication is becoming a key technology for future 6G wireless networks because of its ultra-wide band. However, the implementation of THz communication systems confronts formidable challenges, notably beam splitting effects and high computational complexity associated with them. Our primary objective is to design a hybrid precoder that minimizes the Euclidean distance from the fully digital precoder. The analog precoding part adopts the delay-phase alternating minimization (DP-AltMin) algorithm, which divides the analog precoder into phase shifters and time delayers. This effectively addresses the beam splitting effects within THz communication by incorporating time delays. The traditional digital precoding solution, however, needs matrix inversion in THz massive multiple-input multiple-output (MIMO) communication systems, resulting in significant computational complexity and complicating the design of the analog precoder. To address this issue, we exploit the characteristics of THz massive MIMO communication systems and construct the digital precoder as a product of scale factors and semi-unitary matrices. We utilize Schatten norm and Hölder's inequality to create semi-unitary matrices after initializing the scale factors depending on the power allocation. Finally, the analog precoder and digital precoder are alternately optimized to obtain the ultimate hybrid precoding scheme. Extensive numerical simulations have demonstrated that our proposed algorithm outperforms existing methods in mitigating the beam splitting issue, improving system performance, and exhibiting lower complexity. Furthermore, our approach exhibits a more favorable alignment with practical application requirements, underlying its practicality and efficiency.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2214-2229
• /
• 2024

Wireless sensor network (WSN) consists of large number of sensor nodes that are deployed in geographical locations to collect sensed information, process data and communicate it to the control station for further processing. Due the unfriendly environment where the sensors are deployed, there exist many possibilities of malicious nodes which performs malicious activities in the network. Therefore, the security threats affect performance and life time of sensor networks, whereas various security aspects are there to address security issues in WSN namely Cryptography, Trust Management, Intrusion Detection System (IDS) and Intrusion Prevention Systems (IPS). However, IDS detect the malicious activities and produce an alarm. These malicious activities exploit vulnerabilities in the network layer and affect all layers in the network. Existing feature selection methods such as filter-based methods are not considering the redundancy of the selected features and wrapper method has high risk of overfitting the classification of intrusion. Due to overfitting, the classification algorithm fails to detect the intrusion in better manner. The main objective of this paper is to provide the efficient feature selection algorithm which was suitable for any type classification algorithm to detect the intrusion in an effective manner. This paper, the security of the network is addressed by proposing Feature Selection Algorithm using Chi Squared with Ensemble Method (FSChE). The proposed scheme employs the combination of decision tree along with the random forest classification algorithm to form ensemble classifier. The experimental results justify the feasibility of the proposed scheme in terms of attack detection, packet delivery ratio and time analysis by employing NSL KDD cup data Set. The obtained results shows that the proposed ensemble method increases the overall performance by 10% to 25% with respect to mentioned parameters.

• Journal of Intelligence and Information Systems
• /
• v.21 no.3
• /
• pp.117-129
• /
• 2015

As the Internet becomes ubiquitous, a large volume of information is posted on the Internet with exponential growth every day. Accordingly, it is not unusual that investors in stock markets gather and compile firm-specific or market-wide information through online searches. Importantly, it becomes easier for investors to acquire value-relevant information for their investment decision with the help of powerful search tools on the Internet. Our study examines whether or not the Internet helps investors assess a firm's value better by using firm-level data over long periods spanning from January 2004 to December 2013. To this end, we construct weekly-based search volume for information technology (IT) services firms on the Internet. We limit our focus to IT firms since they are often equipped with intangible assets and relatively less recognized to the public which makes them hard-to measure. To obtain the information on those firms, investors are more likely to consult the Internet and use the information to appreciate the firms more accurately and eventually improve their investment decisions. Prior studies have shown that changes in search volumes can reflect the various aspects of the complex human behaviors and forecast near-term values of economic indicators, including automobile sales, unemployment claims, and etc. Moreover, search volume of firm names or stock ticker symbols has been used as a direct proxy of individual investors' attention in financial markets since, different from indirect measures such as turnover and extreme returns, they can reveal and quantify the interest of investors in an objective way. Following this line of research, this study aims to gauge whether the information retrieved from the Internet is value relevant in assessing a firm. We also use search volume for analysis but, distinguished from prior studies, explore its impact on return comovements with market returns. Given that a firm's returns tend to comove with market returns excessively when investors are less informed about the firm, we empirically test the value of information by examining the association between Internet searches and the extent to which a firm's returns comove. Our results show that Internet searches are negatively associated with return comovements as expected. When sample is split by the size of firms, the impact of Internet searches on return comovements is shown to be greater for large firms than small ones. Interestingly, we find a greater impact of Internet searches on return comovements for years from 2009 to 2013 than earlier years possibly due to more aggressive and informative exploit of Internet searches in obtaining financial information. We also complement our analyses by examining the association between return volatility and Internet search volumes. If Internet searches capture investors' attention associated with a change in firm-specific fundamentals such as new product releases, stock splits and so on, a firm's return volatility is likely to increase while search results can provide value-relevant information to investors. Our results suggest that in general, an increase in the volume of Internet searches is not positively associated with return volatility. However, we find a positive association between Internet searches and return volatility when the sample is limited to larger firms. A stronger result from larger firms implies that investors still pay less attention to the information obtained from Internet searches for small firms while the information is value relevant in assessing stock values. However, we do find any systematic differences in the magnitude of Internet searches impact on return volatility by time periods. Taken together, our results shed new light on the value of information searched from the Internet in assessing stock values. Given the informational role of the Internet in stock markets, we believe the results would guide investors to exploit Internet search tools to be better informed, as a result improving their investment decisions.

• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.1-21
• /
• 2012

In recent years, mobile phone has experienced an extremely fast evolution. It is equipped with high-quality color displays, high resolution cameras, and real-time accelerated 3D graphics. In addition, some other features are includes GPS sensor and Digital Compass, etc. This evolution advent significantly helps the application developers to use the power of smart-phones, to create a rich environment that offers a wide range of services and exciting possibilities. To date mobile AR in outdoor research there are many popular location-based AR services, such Layar and Wikitude. These systems have big limitation the AR contents hardly overlaid on the real target. Another research is context-based AR services using image recognition and tracking. The AR contents are precisely overlaid on the real target. But the real-time performance is restricted by the retrieval time and hardly implement in large scale area. In our work, we exploit to combine advantages of location-based AR with context-based AR. The system can easily find out surrounding landmarks first and then do the recognition and tracking with them. The proposed system mainly consists of two major parts-landmark browsing module and annotation module. In landmark browsing module, user can view an augmented virtual information (information media), such as text, picture and video on their smart-phone viewfinder, when they pointing out their smart-phone to a certain building or landmark. For this, landmark recognition technique is applied in this work. SURF point-based features are used in the matching process due to their robustness. To ensure the image retrieval and matching processes is fast enough for real time tracking, we exploit the contextual device (GPS and digital compass) information. This is necessary to select the nearest and pointed orientation landmarks from the database. The queried image is only matched with this selected data. Therefore, the speed for matching will be significantly increased. Secondly is the annotation module. Instead of viewing only the augmented information media, user can create virtual annotation based on linked data. Having to know a full knowledge about the landmark, are not necessary required. They can simply look for the appropriate topic by searching it with a keyword in linked data. With this, it helps the system to find out target URI in order to generate correct AR contents. On the other hand, in order to recognize target landmarks, images of selected building or landmark are captured from different angle and distance. This procedure looks like a similar processing of building a connection between the real building and the virtual information existed in the Linked Open Data. In our experiments, search range in the database is reduced by clustering images into groups according to their coordinates. A Grid-base clustering method and user location information are used to restrict the retrieval range. Comparing the existed research using cluster and GPS information the retrieval time is around 70~80ms. Experiment results show our approach the retrieval time reduces to around 18~20ms in average. Therefore the totally processing time is reduced from 490~540ms to 438~480ms. The performance improvement will be more obvious when the database growing. It demonstrates the proposed system is efficient and robust in many cases.

• Journal of the Korea Computer Graphics Society
• /
• v.16 no.1
• /
• pp.21-27
• /
• 2010

Today the lectures are usually practiced in a teacher-led traditional classroom system or a student-led e-learning system. Students passively follow the teacher's lectures in both systems, though. Also due to the advances in 3D Computer Graphics and Game technologies, there are trials to exploit the positive effect of games in learning. The serious games, specifically designed games for an educational goal, or existing games for a special class have been used as lectures. Still these games have a great difficulty in being integrated into the educational system technically and economically. Therefore a new 3D MMORPG based lecturing system is presented in this paper. In our new lecturing system, the characteristics of a 3D MMORPG, achievement, sociality, and immersion, are provided to motivate students to participate actively in a lecture. A teacher and students interact with each other in realtime as 3D characters in a 3D virtual classroom on-line. An ordinary teacher can also easily apply our new system to existing classes since a teacher only needs to specify a slide file to prepare a lecture. For the future work, a user study and the effect of our new lecturing system will be performed.

• 한국균학회소식:학술대회논문집
• /
• 2015.05a
• /
• pp.55-55
• /
• 2015

Rust is one of the most destructive diseases on economically important plants such as agricultural and horticultural crops, as well as forest trees [1]. Chemical treatment is the most effective means to control rust, but use of the chemical fungicides involves inevitable risks to human health and environment [2]. Unfortunately, biocontrol is currently impracticable for rust disease management [3]. It is necessary to exploit biocontrol agents to help prevent rust diseases. As a fundamental research for future development of biocontrol agents for rusts, biodiversity of hyperparasites occurring on rust fungi was investigated. During 2006-2010, 197 fungal isolates of the rust hyperparasites were collected and isolated from various combinations of mycohosts and plant hosts in many regions of Korea. Based on morphological and molecular data, they were identified as 8 genera and 12 species. Besides, phylogenetic relationships between the hyperparasites and related taxa were inferred. A total of 114 isolates of Pseudovirgaria were obtained from rust pustules of Phragmidium spp. and Pucciniastrum agrimoniae infecting rosaceous plants. Phylogenetic analysis using multigene sequences revealed a high level of genetic variability among many isolates of Pseudovirgaria and close correlation between the isolates and mycohosts. Only two species of Pseudovirgaria, P. hyperparasitica and P. grisea are often difficult to distinguish by their morphological similarity, but on the molecular basis they were clearly differentiated from each other. There had been no previous record of P. grisea outside Europe, but the present study has proved its presence in Korea. Among six distinct groups (five of P. hyperparasitica and one of P. grisea) within the Pseudovirgaria isolates, each lineage of P. hyperparasitica was closely associated with specific mycohosts and thus might have cospeciated with their mycohosts, which probably led to coevolution. Although P. grisea possesses a host preference for Phragmidium species occurring on Rubus, it was not specific for a mycohost. P. grisea seems to evolve in the direction of having a broad mycohost range. Seventeen isolates of Verticillium-like fungi were isolated from rust sori. Based on morphological data and DNA sequence analysis, the isolates were identified as three Lecanicillium species, viz. L. attenuatum, Lecanicillium sp. 1, Lecanicillium sp. 2, and V. epiphytum. The unidenified two species of Lecanicillium appear to be previously unknown taxa. Sixty-six isolates of miscellaneous hyphomycetes belonging to 6 species of 5 genera were obtained from pustules of rust fungi. On the basis of morphological and molecular analyses, the miscellaneous hyphomycetes growing on rusts were identified as Acrodontium crateriforme, Cladophialophora pucciniophila, Cladosporium cladosporioides, Phacellium vossianum, Ramularia coleosporii, and R. uredinicola.