• Smart Media Journal
• /
• v.11 no.4
• /
• pp.30-37
• /
• 2022

Disease threatens plant growth and recognizing the type of disease is essential to making a remedy. In recent years, deep learning has witnessed a significant improvement for this task, however, a large volume of labeled images is one of the requirements to get decent performance. But annotated images are difficult and expensive to obtain in the agricultural field. Therefore, designing an efficient and effective strategy is one of the challenges in this area with few labeled data. Transfer learning, assuming taking knowledge from a source domain to a target domain, is borrowed to address this issue and observed comparable results. However, current transfer learning strategies can be regarded as a supervised method as it hypothesizes that there are many labeled images in a source domain. In contrast, unsupervised transfer learning, using only images in a source domain, gives more convenience as collecting images is much easier than annotating. In this paper, we leverage unsupervised transfer learning to perform plant disease recognition, by which we achieve a better performance than supervised transfer learning in many cases. Besides, a vision transformer with a bigger model capacity than convolution is utilized to have a better-pretrained feature space. With the vision transformer-based unsupervised transfer learning, we achieve better results than current works in two datasets. Especially, we obtain 97.3% accuracy with only 30 training images for each class in the Plant Village dataset. We hope that our work can encourage the community to pay attention to vision transformer-based unsupervised transfer learning in the agricultural field when with few labeled images.

• Geomechanics and Engineering
• /
• v.28 no.4
• /
• pp.397-404
• /
• 2022

Accurate prediction of the undrained shaft resistance is essential for robust design of bored piles in undrained condition. The undrained shaft resistance is calculated using the undrained adhesion factor multiplied by the undrained cohesion of the soil. However, the available correlations to predict the undrained adhesion factor have been developed using simple regression techniques and the accuracy of these correlations has not been thoroughly assessed in previous studies. The lack of the assessment of these correlations made it difficult for geotechnical engineers to select the most accurate correlation in routine designs. Furthermore, limited attempts have been made in previous studies to use advanced data mining techniques to develop simple and accurate correlation to predict the undrained adhesion factor. This research, therefore, has been conducted to fill these gaps in knowledge by developing novel and robust correlation to predict the undrained adhesion factor. The development of the new correlation has been conducted using the multi-objective evolutionary polynomial regression analysis. The new correlation outperformed the available empirical correlations, where the new correlation scored lower mean absolute error, mean square error, root mean square error and standard deviation of measured to predicted adhesion factor, and higher mean, a20-index and coefficient of correlation. The correlation also successfully showed the influence of the undrained cohesion and the effective stress on the adhesion factor. Hence, the new correlation enhances the design accuracy and can be used by practitioner geotechnical engineers to ensure optimized designs of bored piles in undrained conditions.

• Journal of the Korean earth science society
• /
• v.44 no.1
• /
• pp.79-89
• /
• 2023

The purpose of this study was to investigate the effects of engineering design-based ocean cleanup classes on STEAM attitudes and creative engineering problem-solving dispositions. Furthermore, during this process, we tried to determine interesting points that students encountered in engineering design-based classes. For this study, a science class with six lessons based on engineering design was developed and reviewed by a professor who majored in engineering design, along with five engineering design experts with a master's degree or higher. The subject of the class was selected as the design and implementation of scientific and engineering measures to reduce marine pollution based on the method implemented in an actual Ocean Clean-up Project. The engineering design process utilized the engineering design model presented by NGSS (2013), and was configured to experience redesign through the optimization process. To verify effectiveness, the STEAM attitude questionnaire developed by Park et al. (2019) and the creative engineering problemsolving propensity test tool developed by Kang and Nam (2016) were used. A pre and post t-test was used for statistical analysis for the effectiveness test. In addition, the contents of interesting points experienced by the learners were transcribed after receiving descriptive responses, and were analyzed and visualized through degree centrality analysis. Results confirmed that engineering design in science classes had a positive effect on both STEAM attitude and creative engineering problem-solving disposition (p< .05). In addition, as a result of unstructured data analysis, science and engineering knowledge, engineering experience, and cooperation and collaboration appeared as factors in which learners were interested in learning, confirming that engineering experience was the main factor.

• Journal of Internet Computing and Services
• /
• v.17 no.5
• /
• pp.151-160
• /
• 2016

The purpose of this research is to investigate Korean popular attitudes and social perceptions of 'sharing economy' terminology at the current moment from a creative or socio-economic point of view. In Korea, this study discovers and interprets the objective and tangible annual changes and patterns of sociocultural collective intelligence that have taken place over the last five years by applying text mining in the big data analysis approach. By crawling and Googling, this study collected a significant amount of time series web meta-data with regard to the theme of the sharing economy on the world wide web from 2010 to 2014. Consequently, huge amounts of raw data concerning sharing economy are processed into the value-added meaningful 'word clouding' form of graphs or figures by using the function of word clouding with R programming. Till now, the lack of accumulated data or collective intelligence about sharing economy notwithstanding, it is worth nothing that this study carried out preliminary research on conducting a time-series big data analysis from the perspective of knowledge management and processing. Thus, the results of this study can be utilized as fundamental data to help understand the academic and industrial aspects of future sharing economy-related markets or consumer behavior.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.11
• /
• pp.2874-2890
• /
• 1997

This paper describes a methodology for translation of concurrent software expressed in operating system (OS) calls to Ada. Concurrency is expressed in some legacy software by OS calls that perform concurrent process/task control. Examples considered in this paper are calls in programs in C to Unix and calls in programs in CMS-2 to the Executive Service Routines of ATES or SDEX-20 other software re/reverse engineering research has focused on translating the OS calls in a legacy software to calls to another OS. In this approach, the understanding of software has required knowledge of the underlying OS, which is usually very complicated and informally documented. The research in this paper has focused on translating the OS calls in a legacy software into the equivalent protocols using the Ada facilities. In translation to Ada, these calls are represented by Ada equivalent code that follow the scheme of a message-based kernel oriented architecture. To facilitate translation, it utilizes templates placed in library for data structures, tasks, procedures, and messages. This methodology is a new approach to modeling OS in Ada in software re/reverse engineering. There is no need of knowledge of the underlying OS for software understanding in this approach, since the dependency on the OS in the legacy software is removed. It is portable and interoperable on Ada run-time environments. This approach can handle the OS calls in different legacy software systems.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.2
• /
• pp.154-159
• /
• 2007

Wake distribution data of stem flow fields have been accumulated systematically by model tests. If the correlation between geometrical hull information and wake distribution is grasped through the accumulated data, this correlation can be helpful to designing similar ships. In this paper, Neuro-Fuzzy system that is emerging as a new knowledge over a wide range of fields nowadays is tried to estimate the wake distribution on the propeller plan. Neuro-Fuzzy system is well known as one of prospective and representative analysis method for prediction, classification, diagnosis of real complicated world problem, and it is widely applied even in the engineering fields. For this study three-dimensional stern hull forms and nominal wake values from a model test ate structured as processing elements of input and output layer, respectively. The proposed method is proved as an useful technique in ship design by comparing measured wake distribution with predicted wake distribution.

• 한국전산유체공학회:학술대회논문집
• /
• 2009.04a
• /
• pp.19-19
• /
• 2009

Dynamic Stall is a flow phenomenon which occurs on the retreating side of helicopter rotor blades during forward flight. It also occurs on blades of stall regulated wind turbines under yawing conditions as well as during gust loads. Time scales occurring during this process are comparable on both helicopter and wind turbine blades. Dynamic Stall limits the speed of the helicopter and its manoeuvrability and limits the amount of power production of wind turbines. Extensive numerical as well as experimental investigations have been carried out recently to get detailed insight into the very complex flow structures of the Dynamic Stall process. Numerical codes have to be based on the full equations, i.e. the Navier-Stokes equations to cover the scope of the problems involved: Time dependent flow, unsteady flow separation, vortex development and shedding, compressibility effects, turbulence, transition and 3D-effects, etc. have to be taken into account. In addition to the numerical treatment of the Dynamic Stall problem suitable wind tunnel experiments are inevitable. Comparisons of experimental data with calculated results show us the state of the art and validity of the CFD-codes and the necessity to further improve calculation procedures. In the present paper the phenomenon of Dynamic Stall will be discussed first. This discussion is followed by comparisons of some recently obtained experimental and numerical results for an oscillating helicopter airfoil under Dynamic Stall conditions. From the knowledge base of the Dynamic Stall Problems, the next step can be envisaged: to control Dynamic Stall. The present discussion will address two different Dynamic Stall control methodologies: the Nose-Droop concept and the application of Leading Edge Vortex Generators (LEVoG's) as examples of active and passive control devices. It will be shown that experimental results are available but CFD-data are only of limited comparison. A lot of future work has to be done in CFD-code development to fill this gap. Here mainly 3D-effects as well as improvements of both turbulence and transition modelling are of major concern.

• Journal of Digital Convergence
• /
• v.18 no.5
• /
• pp.75-87
• /
• 2020

The present study determined new value that can be created through the convergence between artificial intelligence technology (AIT) and all industries by deriving and thoroughly analyzing major issues related to artificial intelligence (AI). This study analyzes domestic articles related to AI using topic modeling method based on LDA algorithm. Keywords were extracted from 3,889 articles of eleven metropolitan newspapers, eight business newspapers and major broadcasting companies; articles were selected by searching for the keyword "artificial intelligence". Keywords were extracted by optimizing the relevance parameter λ to improve the measure of pointwise mutual information (PMI), which shows the association among the keywords of each topic, and topic names were inferred from keywords based on valid evidence. The extracted topics widely showed changes occurring throughout society, economy, industries, culture, and the support policy and vision of the government.

• Journal of the Korean Society of Safety
• /
• v.24 no.3
• /
• pp.71-78
• /
• 2009

During a learning process, a human being is assumed to experience knowledge-based behaviors, rule-based behaviors, and skill-based behaviors sequentially if Rasmussen was right. If any psycho-physiological symptom to those different levels can be obtained, it can be useful as a measure whether a human being is fully trained and has gotten a skill in his work. Therefore, this study aimed to draw relationships between human performance measures and psycho-physiological measures while committing a computer-simulated pointing task by utilizing the power spectrum technique of EEG data, especially with the ratio of relative beta-to-alpha band power. The result showed that, during correct responses, the ratio came to stabilize as all the performance data went stable. However, response time was not a simple linear function of task difficulty level only, but a joint function of task characteristics as well as behavior levels. Comparing relative band power ratios from errors and correct responses, activated states of one's brain could be explained, and characteristics of the task could understood. To tell that of pointing task, correlations around C3, C4, P3, P4 and 01, 02 area were significant and high in correct response cases whereas most correlation coefficients went down in error cases standing for imbalance of psycho-motor functions. Though task difficulty was the only one factor that could influence on relative band power ratio with statistical significance, it should be comprehended to mean a different way of expression indicating task characteristics since at least error-some situation could be explained with the help of relative band power ratio that absolute band power failed.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.10a
• /
• pp.406-409
• /
• 2008

We consider the problem of exploiting a taxonomy of propositionalized attributes in order to learn compact and robust classifiers. We introduce Propositionalized Attribute Taxonomy guided Naive Bayes Learner (PAT-NBL), an inductive learning algorithm that exploits a taxonomy of propositionalized attributes as prior knowledge to generate compact and accurate classifiers. PAT-NBL uses top-down and bottom-up search to find a locally optimal cut that corresponds to the instance space from propositionalized attribute taxonomy and data. Our experimental results on University of California-Irvine (UCI) repository data sets show that the proposed algorithm can generate a classifier that is sometimes comparably compact and accurate to those produced by standard Naive Bayes learners.