• The Journal of the Convergence on Culture Technology
• /
• v.10 no.1
• /
• pp.517-524
• /
• 2024

Generative AI, exemplified by models like ChatGPT, has revolutionized human-machine interactions in the 21st century. As these advancements permeate various sectors, their intersection with the arts is both promising and challenging. Despite the arts' historical resistance to AI replacement, recent developments have sparked active research in AI's role in artistry. This study delves into the potential of AI in visual arts education, highlighting the necessity of swift adaptation amidst the Fourth Industrial Revolution. This research, conducted at a 4-year global higher education institution located in Gyeongbuk, involved 70 participants who took part in a creative convergence module course project. The study aimed to examine the influence of AI collaboration in visual arts, analyzing distinctions across majors, grades, and genders. The results indicate that creative activities with AI positively influence students' creativity and digital media literacy. Based on these findings, there is a need to further develop effective educational strategies and directions that incorporate AI.

• Korean Journal of Applied Biomechanics
• /
• v.25 no.2
• /
• pp.175-182
• /
• 2015

Objective : The purpose of this study was to investigate difference in fascicle behavior of the medial gastrocnemius during the locomotion with varying intensities, such as gait and one-legged and two-legged vertical jumping. Methods : Six subjects (3 males and 3 females; age: $27.2{\pm}1.6yrs.$, body mass: $62.8{\pm}9.8kg$, height: $169.6{\pm}8.5cm$) performed normal gait (G) at preferred speed and maximum vertical jumping with one (OJ) and two (TJ) legs. While subjects were performing the given tasks, the hip, knee and ankle joint motion and ground reaction force was monitored using a 8-infrared camera motion analysis system with two forceplates. Simultaneously, electromyography of the triceps surae muscles, and the fascicle length of the medial gastrocnemius were recorded using a real-time ultrasound imaging machine. Results : Comparing to gait, the kinematic and kinetic parameters of TJ and OJ were found to be significantly different. Along with those parameters, change in the medial gastrocnemius (MG) muscle-tendon complex (MTC) length ($50.57{\pm}6.20mm$ for TJ and $44.14{\pm}5.39mm$ for OJ) and changes in the fascicle length of the MG ($18.97{\pm}3.58mm$ for TJ and $20.31{\pm}4.59mm$ for OJ) were observed. Although the total excursion of the MTC and the MG fascicle length during the two types of jump were not significantly different, however the pattern of length changes were found to be different. For TJ, the fascicle length maintained isometric longer during the propulsive phase than OJ. Conclusion : One-legged and two-legged vertical jumping use different muscle-tendon interaction strategies.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.

• Restorative Dentistry and Endodontics
• /
• v.33 no.2
• /
• pp.90-97
• /
• 2008

The purpose of this study was to evaluate the effect of passive or active application of primer and coat times of bond on the shear bond strength when a self-etching primer adhesive (Clearfil SE Bond) was applied to enamel surface. Crowns of sixteen human molars were selected. Buccal and lingual enamels of crowns were partially exposed and slabs of 1.2 mm thick were made. They were divided into one of four equal groups (n = 8). Group 1: passive application of Primer and 1 coat of Bond, Group 2: active application of Primer and 1 coat of Bond, Group 3: passive application of Primer and 2 coats of Bond, Group 4: active application of Primer and 2 coats of Bond. Clearfil AP-X was bonded to enamel suface of each group using Tygon tubes. The bonded specimens were subjected to microshear bond strength (uSBS) testing with a crosshead speed of 1 mm/min. The results of this study were as follows; 1. The uSBS of Group 1 was the lowest among groups and the uSBS of Group 4 was the highest. 2. There was not statistically significant interaction between enamel uSBS by application method of Primer and coat time of Bond (p > 0.05). 3. There was not statistically significant difference between enamel uSBS by passive and active application of Primer (p > 0.05). 4. There was statistically significant difference between enamel uSBS by one- and two-coat of Bond (p < 0.05).

• Korean Journal of Cognitive Science
• /
• v.21 no.4
• /
• pp.447-458
• /
• 2010

Time expressions are a very important form of information in different types of data. Thus, the recognition of a time expression is an important factor in the field of information extraction. However, most previously designed systems consider only a specific domain, because time expressions do not have a regular form and frequently include different ellipsis phenomena. We present a two-level recognition method consisting of extraction and transformation phases to achieve generality and portability. In the extraction phase, time expressions are extracted by atomic time units for extensibility. Then, in the transformation phase, omitted information is restored using basis time and prior knowledge. Finally, every complete atomic time unit is transformed into a normalized form. The proposed system can be used as a general-purpose system, because it has a language- and domain-independent architecture. In addition, this system performs robustly in noisy data like SMS data, which include various errors. For SMS data, the accuracies of time-expression extraction and time-expression normalization by using the proposed system are 93.8% and 93.2%, respectively. On the basis of these experimental results, we conclude that the proposed system shows high performance in noisy data.

• Proceedings of the Korean Society for Applied Microbiology Conference
• /
• 2001.06a
• /
• pp.83-89
• /
• 2001

Recent advances in the structural and molecular biology uncovered that a set of translation factors resembles a tRNA shape and, in one case, even mimics a tRNA function for deciphering the genetic ：ode. Nature must have evolved this 'art' of molecular mimicry between protein and ribonucleic acid using different protein architectures to fulfill the requirement of a ribosome 'machine'. Termination of protein synthesis takes place on the ribosomes as a response to a stop, rather than a sense, codon in the 'decoding' site (A site). Translation termination requires two classes of polypeptide release factors (RFs)： a class-I factor, codon-specific RFs (RFI and RF2 in prokaryotes; eRFI in eukaryotes), and a class-IT factor, non-specific RFs (RF3 in prokaryotes; eRF3 in eukaryotes) that bind guanine nucleotides and stimulate class-I RF activity. The underlying mechanism for translation termination represents a long-standing coding problem of considerable interest since it entails protein-RNA recognition instead of the well-understood codon-anticodon pairing during the mRNA-tRNA interaction. Molecular mimicry between protein and nucleic acid is a novel concept in biology, proposed in 1995 from three crystallographic discoveries, one, on protein-RNA mimicry, and the other two, on protein-DNA mimicry. Nyborg, Clark and colleagues have first described this concept when they solved the crystal structure of elongation factor EF- Tu：GTP：aminoacyl-tRNA ternary complex and found its overall structural similarity with another elongation factor EF-G including the resemblance of part of EF-G to the anticodon stem of tRNA (Nissen et al. 1995). Protein mimicry of DNA has been shown in the crystal structure of the uracil-DNA glycosylase-uracil glycosylase inhibitor protein complex (Mol et al. 1995; Savva and Pear 1995) as well as in the NMR structure of transcription factor TBP-TA $F_{II}$ 230 complex (Liu et al. 1998). Consistent with this discovery, functional mimicry of a major autoantigenic epitope of the human insulin receptor by RNA has been suggested (Doudna et al. 1995) but its nature of mimic is. still largely unknown. The milestone of functional mimicry between protein and nucleic acid has been achieved by the discovery of 'peptide anticodon' that deciphers stop codons in mRNA (Ito et al. 2000). It is surprising that it took 4 decades since the discovery of the genetic code to figure out the basic mechanisms behind the deciphering of its 64 codons.