• Language and Information
• /
• v.12 no.1
• /
• pp.27-65
• /
• 2008

This paper presents a set of methodologies for aligning hypernym-hyponym noun pairs between Korean and English, based on the EuroWordNet approach. Following the methods conducted in EuroWordNet, our approach makes extensive use of WordNet in four steps of the building process: 1) Monolingual dictionaries have been used to extract proper hypernym-hyponym noun pairs, 2) bilingual dictionary has converted the extracted pairs, 3) Word Net has been used as a backbone of alignment criteria, and 4) WordNet has been used to select the most similar pair among the candidates. The importance of this study lies not only on enriching semantic links between two languages, but also on integrating lexical resources based on a language specific and dependent structure. Our approaches are aimed at building an accurate and detailed lexical resource with proper measures rather than at fast development of generic one using NLP technique.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.185-191
• /
• 2022

In this study, we propose a novel super-resolution neural network based on U-Net, residual neural network, and sub-pixel convolution. To prevent the loss of detailed information due to the max pooling of U-Net, we propose down-sampling and connection using sub-pixel convolution. This uses all pixels in the filter, unlike the max pooling that creates a new feature map with only the max value in the filter. As a 2×2 size filter passes, it creates a feature map consisting only of pixels in the upper left, upper right, lower left, and lower right. This makes it half the size and quadruple the number of feature maps. And we propose two methods to reduce the computation. The first uses sub-pixel convolution, which has no computation, and has better performance, instead of up-convolution. The second uses a layer that adds two feature maps instead of the connection layer of the U-Net. Experiments with a banchmark dataset show better PSNR values on all scale and benchmark datasets except for set5 data on scale 2, and well represent local area patterns.

• Language and Information
• /
• v.14 no.1
• /
• pp.113-144
• /
• 2010

The primary goal of this paper is to find a feasible way to answer the question: Does the similarity in meaning between verbs relate to the similarity in their subcategorization? In order to answer this question in a rather concrete way on the basis of a large set of English verbs, this study made use of various language resources, tools, and statistical methodologies. We first compiled a list of 678 verbs that were selected from the most and second most frequent word lists from the Colins Cobuild English Dictionary, which also appeared in WordNet 3.0. We calculated similarity measures between all the pairs of the words based on the 'jcn' algorithm (Jiang and Conrath, 1997) implemented in the WordNet::Similarity module (Pedersen, Patwardhan, and Michelizzi, 2004). The clustering process followed, first building similarity matrices out of the similarity measure values, next drawing dendrograms on the basis of the matricies, then finally getting 177 meaningful clusters (covering 437 verbs) that passed a certain level set by z-score. The subcategorization frames and their frequency values were taken from the ICE-GB. In order to calculate the Selectional Preference Strength (SPS) of the relationship between a verb and its subcategorizations, we relied on the Kullback-Leibler Divergence model (Resnik, 1996). The SPS values of the verbs in the same cluster were compared with each other, which served to give the statistical values that indicate how much the SPS values overlap between the subcategorization frames of the verbs. Our final analysis shows that the degree of overlap, or the relationship between semantic similarity and the subcategorization frames of the verbs in English, is equally spread out from the 'very strongly related' to the 'very weakly related'. Some semantically similar verbs share a lot in terms of their subcategorization frames, and some others indicate an average degree of strength in the relationship, while the others, though still semantically similar, tend to share little in their subcategorization frames.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.40 no.7
• /
• pp.685-692
• /
• 2016

A cutting head is an attachment on the front of an excavator that cuts or grinds rocks. Cutting tools, called pick cutters, are arranged on the surface of the cutting head. The exact arrangement and configuration of pick cutters is one of the most important factors in determining grinding efficiency. This study focuses on the problem of automation for pick arrangement design, in order to make the design process more efficient and convenient. Design automation was carried out using RecurDyn/ProcessNet, and it was composed of three parts: 'Drum set', 'Pick load', and 'Pick arrangement' sections. The presented method helps to decrease costs attributed to designing cutting heads and can be used to generate a wide range of attachment mechanisms.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.33 no.2
• /
• pp.109-117
• /
• 1997

In order to hold the behavior of fish school to the set-net, a series of tag-recapture experiments were carried out in two fishing grounds of the middle sized set-nets which were located in 20m depth on the coast of Keojedo and Namhaedo in the Southern part of Korea from September to October in 1996. In the experiments, the leading ability of the leader and the fish court and the recapturing ability of the bag nets were checked out for the six species of fish in method of discharging the tagged fishes at side points of leader, and the middle points of the fish court and the bag nets in a hauling step, and recapturing them at the bag nets in the next hauling. The results obtained are as follows; 1. The ratio recaptured at the both side bag-nets in the next hauling after discharged from the fish court in the previous step was 20.3% in small size of mackerel Scomber japonicus, 16.2% in small size of horse mackerel Trachurus japonicus, 10.3% in black sea-bream Acanthopagrus schlegelii, 19.1% in red barracuda Syhyraena pinguis, 16.3% in halfbeak Hemiramphus sajori, 20.0% in gizzard shad Konosirus punctatus individually, and totally in six species of fish, that was 17.2%. 2. The ratio recaptured at the same bag net after discharged in the both side bag-nets was 21.7% in small size of mackerel, 21.5% in small size of horse mackerel, 6.7% in black sea-bream, 17.8% in red barracuda, 16.8 in half-beak, 19.1% in gizzard shad individually, and totally in six species of fish, that was 18.8%. 3. The leading ratio from side points of the leader departed from door in 25m to fish court was 58.9% in small size of mackerel, 74.6% in small size of horse mackerel, 38.0% in black sea-bream, 54.7% in red barracuda, 58.6% in half-beak, 54.5% in gizzard shad individually, and totally in six species of fish, that was 57.8%. So it was assumed that the leader of set-net was very effective in leading to the swimming direction of small size of mackerel, small size of horse mackerel, red barracuda, half-beak and gizzard shad. 4. Red barracuda, half-beak and gizzard shad entered into bag net of upstream in large numbers than bag net of downstream, and small size of horse mackerel and black sea-bream entered into bag net of downstream in large numbers than bag net of upstream. 5. Small size of mackerel and small size of horse mackerel had high remaining rate in the bag net of downstream, and black sea-bream, red barracuda and half-beak had high remaining rate in the bag net of upstream.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.27 no.1
• /
• pp.31-40
• /
• 1991

This paper described on the availability fo the underwater telemetering by the ulterasonic multi-beam system made as a trial to expand detectable range of the fish school. The ultrasonic multi-beam system consisted of four transducers which reconstructed with the existing net recorder. The experiment for the telemetering carried out in the set net fishing ground. The results obtained are summerized as follows: 1. The detectable distance of a target by the linear arrangement of four transducers increased according to the sea depth and the interval between transducers. 2. When the fish school in the entrance of set net was measured by linear arrangement of transducers it was entered in depth of 2.5~3.5m at near position of leader, and in depth of 3.5~4.5m at near position of door net. 3. The deviations of error between the actual position and the position by transducer in case of the target depth 1m, 1.5m, 2m were 5.9~27.1cm, 3.2~28.9cm, 3.5~25.8cm respectively, and 68.3% probability radius of them were 14.6cm, 17.7cm, 17.0cm respectively. 4. When the fish school in the fish court of set net was measured by plane arrangement of transducer it was entered toward the opposite direction of tide current. 5. The available distance of telemetering by the multi-beam transducer was 1.8km and the telemetering was possible to control everywhere in case of sea depth more than three meters.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.398-400
• /
• 2019

Faced with the increasing demand for image recognition on mobile devices, how to run convolutional neural network (CNN) models on mobile devices with limited computing power and limited storage resources encourages people to study efficient model design. In recent years, many effective architectures have been proposed, such as mobilenet_v1, mobilenet_v2 and PeleeNet. However, in the process of feature selection, all these models neglect some information of shallow features, which reduces the capture of shallow feature location and semantics. In this study, we propose an effective framework based on Feature Pyramid Networks to improve the recognition accuracy of deep and shallow images while guaranteeing the recognition speed of PeleeNet structured images. Compared with PeleeNet, the accuracy of structure recognition on CIFA-10 data set increased by 4.0%.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.16 no.2
• /
• pp.95-108
• /
• 2021

Technology innovation activities are very important for companies to secure technological competitiveness and continue to grow. Korea operates a certification system at the national level to promote corporate innovation activities, and strives to enhance SMEs' global technological competitiveness. Among these, the representative system related to technological innovation is the New Excellent Technology (NET) certification. NET is certified through a strict three-stage screening process, and is operated for the purpose of commercialization of new technology, technology trading, and promotion of early market entry by companies. Acquiring NET certification means that the company has a certain level of technological competitiveness. Therefore, this study attempted to conduct an empirical analysis on which technology innovation activities of companies affect the success of R&D projects and improvement of management performance, centering on NET certification system. To verify this, technology strategy, technology planning, systematic R&D process, internal cooperation, and external cooperation activities were set as major variables. As a result of the empirical analysis, it was confirmed that all variables set in this study individually contributed to the success of the R&D project and improvement of management performance. However, when looking at a comprehensive level that considers all variables, it was analyzed that systematic R&D process management and cooperation activities with external organizations have a statistically significant effect on R&D project success, and technology strategy establishment and technology planning activities, which are the initial stages of R&D, have a statistically significant effect on management performance. This study was conducted on innovation-oriented SMEs that have established and operated corporate R&D centers and are actively conducting R&D activities, and multiple regression analysis was used as an analysis method.

• Smart Media Journal
• /
• v.11 no.5
• /
• pp.26-37
• /
• 2022

In this paper, images are automatically acquired to control the initial disease of strawberries among facility cultivation crops, and disease analysis is performed using the EfficientNet model to inform farmers of disease status, and disease diagnosis service is proposed by experts. It is possible to obtain an image of the strawberry growth stage and quickly receive expert feedback after transmitting the disease diagnosis analysis results to farmers applications using the learned EfficientNet model. As a data set, farmers who are actually operating facility cultivation were recruited and images were acquired using the system, and the problem of lack of data was solved by using the draft image taken with a cell phone. Experimental results show that the accuracy of EfficientNet B0 to B7 is similar, so we adopt B0 with the fastest inference speed. For performance improvement, Fine-tuning was performed using a pre-trained model with ImageNet, and rapid performance improvement was confirmed from 100 Epoch. The proposed service is expected to increase production by quickly detecting initial diseases.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.5
• /
• pp.161-168
• /
• 2021

Cracks affect the robustness of infrastructures such as buildings, bridge, pavement, and pipelines. This paper presents an automated crack detection system which detect cracks in diverse surfaces. We first constructed the combined crack dataset, consists of multiple crack datasets in diverse domains presented in prior studies. Then, state-of-the-art deep learning models in computer vision tasks including VGG, ResNet, WideResNet, ResNeXt, DenseNet, and EfficientNet, were used to validate the performance of crack detection. We divided the combined dataset into train (80%) and test set (20%) to evaluate the employed models. DenseNet121 showed the highest accuracy at 96.20% with relatively low number of parameters compared to other models. Based on the validation procedures of the advanced deep learning models in crack detection task, we shed light on the cost-effective automated crack detection system which can be applied to different surfaces and structures with low computing resources.