• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.13-20
• /
• 2022

It's proposed and analyzed ML(Machine Learning) models to predict vehicle FC(Fuel Consumption) in real-time. The test driving was done for a car to measure vehicle speed, acceleration, road gradient and FC for training dataset. The various ML models were trained with feature data of speed, acceleration and road-gradient for target FC. There are two kind of ML models and one is regression type of linear regression and k-nearest neighbors regression and the other is classification type of k-nearest neighbors classifier, logistic regression, decision tree, random forest and gradient boosting in the study. The prediction accuracy is low in range of 0.5 ~ 0.6 for real-time FC and the classification type is more accurate than the regression ones. The prediction error for total FC has very low value of about 0.2 ~ 2.0% and regression models are more accurate than classification ones. It's for the coefficient of determination (R²) of accuracy score distributing predicted values along mean of targets as the coefficient decreases. Therefore regression models are good for total FC and classification ones are proper for real-time FC prediction.

• Journal of Broadcast Engineering
• /
• v.27 no.4
• /
• pp.538-547
• /
• 2022

Recently introduced image classification methods using Transformers show remarkable performance improvements over conventional neural network-based methods. In order to effectively consider regional features, research has been actively conducted on how to apply transformers by dividing image areas into multiple window areas, but learning of inter-window relationships is still insufficient. In this paper, to overcome this problem, we propose a transformer structure that can reflect the relationship between windows in learning. The proposed method computes the importance of each window region through compression and a fully connected layer based on self-attention operations for each window region. The calculated importance is scaled to each window area as a learned weight of the relationship between the window areas to re-calibrate the feature value. Experimental results show that the proposed method can effectively improve the performance of existing transformer-based methods.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.3
• /
• pp.521-526
• /
• 2023

Due to the recent pandemic, most educational systems are being conducted through online classes. Unlike face-to-face classes, it is even more difficult for learners to maintain concentration, and evaluating the learners' attitude toward the class is also challenging. In this paper, we proposed a real-time concentration-based review support system for learners in real-time video lectures that can be used in online classes. This system measured the learner's face, pupils, and user activity in real-time using the equipment used in the existing video system, and delivers real-time concentration measurement values to the instructor in various forms. At the same time, if the concentration measurement value falls below a certain level, the system alerted the learner and records the timestamp of the lecture. By using this system, instructors can evaluate the learners' participation in the class in real-time and help to improve their class abilities.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.12
• /
• pp.1786-1793
• /
• 2022

Cyber-attacks such as smishing and hacking mail exploiting COVID-19, political and social issues, have recently been continuous. Machine learning and deep learning technology research are conducted to prevent any damage due to cyber-attacks inducing malicious links to breach personal data. It has been concluded as a lack of basis to judge the attacks to be malicious in previous studies since the features of data set were excessively simple. In this paper, nine main features of three types, "URL Days", "URL Word", and "URL Abnormal", were proposed in addition to lexical features of URL which have been reflected in previous research. F1-Score and accuracy index were measured through four different types of machine learning algorithms. An improvement of 0.9% in a result and the highest value, 98.5%, were examined in F1-Score and accuracy through comparatively analyzing an existing research. These outcomes proved the main features contribute to elevating the values in both accuracy and performance.

• SUVANNABHUMI
• /
• v.8 no.1
• /
• pp.121-146
• /
• 2016

Defining key characteristics of Southeast Asia requires historical interpretation. Southeast Asia is a diverse and complicated region, but some of modern history's "grand narratives" serve to unify its historical experience. At a minimum, the modern history of the region involves decisive encounters with universal religions, the rise of Western colonialism, the experience of world wars, decolonization, and the end of the "cycle of violence". The ability of the region's peoples to adapt to these many challenges and successfully build new nations is a defining feature of Southeast Asia's place in the global stage. This paper will begin with a question: is it possible to develop a hermeneutic of "expedience" as a way to interpret the region's history? That is, rather than regard the region from a purely Western, nationalist, "internalist" point of view, it would be useful to identify a new series of interpretative contexts from which to begin scholarly analysis. In order to contextualize this discussion, the paper will draw upon the writings of figures who explored the region before knowledge about it was shaped by purely colonist or nationalist enterprises. To this end, particular attention will be devoted to exploring some of John Furnivall's ways of conceptualizing Southeast Asia. Investigating Furnivall, a critic of colonialism, will be done in relation to his historical situation. Because Furnivall's ideas have played a pivotal role in the interpretation of Southeast Asia, the paper will highlight the intellectual history of the region in order to ascertain the value of these concepts for subsequent historical interpretation. Ultimately, the task of interpreting the region's history requires a framework which will move beyond the essentializing orientalist categories produced by colonial scholarship and the reactionary nation-building narratives which followed. Instead, by beginning with a mode of historical interpretation that focuses on the many realities of expedience which have been necessary for the region's peoples, it may be possible to write a history which highlights the extraordinarily adaptive quality of Southeast Asia's populations, cultures, and nations. To tell this story, which would at once highlight key characteristics of the region while showing how they developed through historical encounters, would go a long way to capturing Southeast Asia's contribution's to global development.

• Korean Journal of Radiology
• /
• v.24 no.6
• /
• pp.512-521
• /
• 2023

Objective: There is increasing recognition that left atrial (LA) strain can be a prognostic marker of various cardiac diseases. However, its prognostic value in acute myocarditis remains unclear. Therefore, this study aimed to evaluate whether cardiovascular magnetic resonance (CMR)-derived parameters of LA strain can predict outcomes in patients with acute myocarditis. Materials and Methods: We retrospectively analyzed the data of 47 consecutive patients (44.2 ± 18.3 years; 29 males) with acute myocarditis who underwent CMR in 13.5 ± 9.7 days (range, 0-31 days) of symptom onset. Various parameters, including feature-tracked CMR-derived LA strain, were measured using CMR. The composite endpoints included cardiac death, heart transplantation, implantable cardioverter-defibrillator or pacemaker implantation, rehospitalization following a cardiac event, atrial fibrillation, or embolic stroke. The Cox regression analysis was performed to identify associations between the variables derived from CMR and the composite endpoints. Results: After a median follow-up of 37 months, 20 of the 47 (42.6%) patients experienced the composite events. In the multivariable Cox regression analysis, LA reservoir and conduit strains were independent predictors of the composite endpoints, with an adjusted hazard ratio per 1% increase of 0.90 (95% confidence interval [CI], 0.84-0.96; P = 0.002) and 0.91 (95% CI, 0.84-0.98; P = 0.013), respectively. Conclusion: LA reservoir and conduit strains derived from CMR are independent predictors of adverse clinical outcomes in patients with acute myocarditis.

• Geomechanics and Engineering
• /
• v.37 no.1
• /
• pp.49-64
• /
• 2024

Rocks undergoing repeated loading and unloading over an extended period, such as due to earthquakes, human excavation, and blasting, may result in the gradual accumulation of stress and deformation within the rock mass, eventually reaching an unstable state. In this study, a CNN-CCM is proposed to address the mechanical behavior. The structure and hyperparameters of CNN-CCM include Conv2D layers × 5; Max pooling2D layers × 4; Dense layers × 4; learning rate=0.001; Epoch=50; Batch size=64; Dropout=0.5. Training and validation data for deep learning include 71 rock samples and 122,152 data points. The AI Rock Constitutive Model learned by CNN-CCM can predict strain values(ε₁) using Mass (M), Axial stress (σ₁), Density (ρ), Cyclic number (N), Confining pressure (σ₃), and Young's modulus (E). Five evaluation indicators R², MAPE, RMSE, MSE, and MAE yield respective values of 0.929, 16.44%, 0.954, 0.913, and 0.542, illustrating good predictive performance and generalization ability of model. Finally, interpreting the AI Rock Constitutive Model using the SHAP explaining method reveals that feature importance follows the order N > M > σ₁ > E > ρ > σ₃.Positive SHAP values indicate positive effects on predicting strain ε₁ for N, M, σ₁, and σ₃, while negative SHAP values have negative effects. For E, a positive value has a negative effect on predicting strain ε₁, consistent with the influence patterns of conventional physical rock constitutive equations. The present study offers a novel approach to the investigation of the mechanical constitutive model of rocks under cyclic loading and unloading conditions.

• Journal of Microbiology and Biotechnology
• /
• v.33 no.11
• /
• pp.1448-1456
• /
• 2023

A Gram-positive, non-motile, and non-spore-forming lactic acid bacterium, designated as BK2^T, was isolated from kimchi, a Korean traditional fermented vegetable food, and the taxonomic characteristics of strain BK2^T, along with strain LMG 11983, were analyzed. Both strains optimally grew at 30℃, pH 7.0, and 1.0% NaCl. Cells of both strains were heterofermentative and facultatively anaerobic rods, demonstrating negative reactions for catalase and oxidase. Major fatty acids (>10%) identified in both strains were C_18:1 ω9c, C_16:0, and summed feature 7 (comprising C_19:1 ω6c and/or C_19:1 ω7c). The genomic DNA G+C contents of both strains were 44.7 mol%. The 16S rRNA gene sequence similarity (99.9%), average nucleotide identity (ANI; 99.9%), and digital DNA-DNA hybridization (dDDH; 99.7%) value between strains BK2^T and LMG 11983 indicated that they are different strains of the same species. Strain BK2^T was most closely related to Weissella confusa JCM 1093^T and Weissella cibaria LMG 17699^T, with 100% and 99.4% 16S rRNA gene sequence similarities, respectively. However, based on the ANI and dDDH values (92.3% and 48.1% with W. confusa, and 78.4% and 23.5% with W. cibaria), it was evident that strain BK2^T represents a distinct species separate from W. confusa and W. cibaria. Based on phylogenetic, phenotypic, and chemotaxonomic features, strains BK2^T and LMG 11983 represent a novel species of the genus Weissella, for which the name Weissella fermenti sp. nov. is proposed. The type of strain is BK2^T (=KACC 22833^T=JCM 35750^T).

• Journal of Internet Computing and Services
• /
• v.25 no.5
• /
• pp.87-94
• /
• 2024

A collaborative filtering-based recommender system is a method that gives priority to items preferred by similar neighbors when providing recommended items for the current user. The similarity measure is very important for the performance of the system. In this study, a genetic algorithm was used to calculate the similarity value between users that results in optimal performance. In particular, the genetic algorithm was run separately for each rated item feature to improve prediction accuracy performance. Through performance experiments, the optimal probabilities of the genetic algorithm operators were obtained, and as a result of experiments using two types of public datasets, it was confirmed that the prediction performance of the proposed method was superior to that of existing methods, especially in a sparse data environment. The results of this study can improve the accuracy of personalized recommendations and be effectively applied in real-world applications with large-scale user and item data.

• Advanced Industrial SCIence
• /
• v.3 no.3
• /
• pp.8-13
• /
• 2024

Zero Knowledge Proof (ZKP) is an innovative decentralized technology designed to enhance the privacy and security of virtual currency transactions. By ensuring that only the necessary information is disclosed by the transaction provider, ZKP protects the confidentiality of all parties involved. This ensures that both the identity of the transacting parties and the transaction value remain confidential.ZKP not only provides a robust privacy function by concealing the identities and values involved in blockchain transactions but also facilitates the exchange of money between parties without the need to verify each other's identity. This anonymity feature is crucial in promoting trust and security in financial transactions, making ZKP a pivotal technology in the realm of virtual currencies. In the context of the Fourth Industrial Revolution, the application of ZKP contributes significantly to the comprehensive and stable development of financial services. It fosters a trustworthy user environment by ensuring that transaction privacy is maintained, thereby encouraging broader adoption of virtual currencies. By integrating ZKP, financial services can achieve a higher level of security and trust, essential for the continued growth and innovation within the sector.