• KIISE Transactions on Computing Practices
• /
• v.23 no.5
• /
• pp.316-321
• /
• 2017

We analyze the online Hangul handwriting recognition problem (HHR) and present solutions based on recurrent neural networks. The solutions are organized according to the three kinds of sequence labeling problem - sequence classifications, segment classification, and temporal classification, with additional consideration of the structural constitution of Hangul characters. We present a stacked gated recurrent unit (GRU) based model as the natural HHR solution in the sequence classification level. The proposed model shows 86.2% accuracy for recognizing 2350 Hangul characters and 98.2% accuracy for recognizing the six types of Hangul characters. We show that the type recognizing model successfully follows the type change as strokes are sequentially written. These results show the potential for RNN models to learn high-level structural information from sequential data.

• The Korean Journal of Physiology and Pharmacology
• /
• v.7 no.6
• /
• pp.295-301
• /
• 2003

Striatum plays a crucial role in the movement control and habitual learning. It receives an information from wide area of cerebral cortex as well as an extensive serotonergic (5-hydroxytryptamine, 5-HT) input from raphe nuclei. In the present study, the effects of 5-HT to modulate synaptic transmission were studied in the rat corticostriatal brain slice using in vitro extracellular recording technique. Synaptic responses were evoked by stimulation of cortical glutamatergic inputs on the corpus callosum and recorded in the dorsal striatum. 5-HT reversibly inhibited coticostriatal glutamatergic synaptic transmission in a dose-dependent fashion (5, 10, 50, and $10{\mu}M$), maximally reducing in the corticostriatal population spike (PS) amplitude to $40.1{\pm}5.0$% at a concentration of $50{\mu}M$ 5-HT. PSs mediated by non-NMDA glutamate receptors, which were isolated by bath application of the NMDA receptor antagonist, d,l-2-amino-5-phospohonovaleric acid (AP-V), were decreased by application of $50{\mu}M$ 5-HT. However, PSs mediated by NMDA receptors, that were activated by application of zero $Mg^{2+}$ aCSF, were not significantly affected by $50{\mu}M$ 5-HT. To test whether the corticostriatal synaptic inhibitions by 5-HT might involve a change in the probability of neurotransmitter release from presynaptic nerve terminals, we measured the paired-pulse ratio (PPR) evoked by 2 identical pulses (50 ms interpulse interval), and found that PPR was increased ($33.4{\pm}5.2$%) by 5-HT, reflecting decreased neurotransmitter releasing probability. These results suggest that 5-HT may decrease neurotransmitter release probability of glutamatergic corticostriatal synapse and may be able to selectively decrease non-NMDA glutamate receptor-mediated synaptic transmission.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.34-34
• /
• 2023

The Chah-Nimeh reservoirs, which are a sort of natural lakes located in the border of Iran and Afghanistan, are the main drinking and agricultural water resources of Sistan arid region. Considering the occurrence of intense seasonal wind, locally known as levar wind, this study aims to explore the possibility to provide a TSM (Total Suspended Matter) monitoring model of Chah-Nimeh reservoirs using multi-temporal satellite images and in-situ wind speed data. The results show that a strong correlation between TSM concentration and wind speed are present. The developed empirical model indicated high performance in retrieving spatiotemporal distribution of the TSM concentration with R²=0.98 and RMSE=0.92g/m³. Following this observation, we also consider a machine learning-based model to predicts the average TSM using only wind speed. We connect our in-situ wind speed data to the TSM data generated from the inversion of multi-temporal satellite imagery to train a neural network based mode l(Wind2TSM-Net). Examining Wind2TSM-Net model indicates this model can retrieve the TSM accurately utilizing only wind speed (R²=0.88 and RMSE=1.97g/m³). Moreover, this results of this study show tha the TSM concentration can be estimated using only in situ wind speed data independent of the satellite images. Specifically, such model can supply a temporally persistent means of monitoring TSM that is not limited by the temporal resolution of imagery or the cloud cover problem in the optical remote sensing.

• Journal of Bio-Environment Control
• /
• v.26 no.4
• /
• pp.411-417
• /
• 2017

Environmental and growth factors such as light intensity, vapor pressure deficit, and leaf area index are important variables that can change the transpiration rate of plants. The objective of this study was to compare the transpiration rates estimated by modified Penman-Monteith model and artificial neural network. The transpiration rate of paprika (Capsicum annuum L. cv. Fiesta) was obtained by using the change in substrate weight measured by load cells. Radiation, temperature, relative humidity, and substrate weight were collected every min for 2 months. Since the transpiration rate cannot be accurately estimated with linear equations, a modified Penman-Monteith equation using compensated radiation (Shin et al., 2014) was used. On the other hand, ANN was applied to estimating the transpiration rate. For this purpose, an ANN composed of an input layer using radiation, temperature, relative humidity, leaf area index, and time as input factors and five hidden layers was constructed. The number of perceptons in each hidden layer was 512, which showed the highest accuracy. As a result of validation, $R^2$ values of the modified model and ANN were 0.82 and 0.94, respectively. Therefore, it is concluded that the ANN can estimate the transpiration rate more accurately than the modified model and can be applied to the efficient irrigation strategy in soilless cultures.

• Journal of Engineering Education Research
• /
• v.14 no.3
• /
• pp.38-44
• /
• 2011

The purpose of this study is to provide the information of decision making that can be used to improve curriculum of engineering education by surveying and analyzing that educational needs of ${\bigcirc}{\bigcirc}$ university' engineering students about program outcomes. The conclusions of this study are as follows. First, it was found that engineering students surveyed valued much of the necessity of PO2(analysis experiment), PO3(design capability), PO6(teamwork) among program outcomes. In addition, it was found that engineering students surveyed thought their ability was low in PO3(design capability), PO4(problem solving), PO10 (knowledge of current events). Second, it was found that the order of educational needs about program outcomes was PO3(design capability), PO2(analysis experiment), PO1(knowledge application), PO5(practical ability), PO4(problem solving) which suggested that engineering students surveyed had high educational needs for engineering program outcomes. On the other hand, it was found that engineering students surveyed showed lower awareness of PO7, PO10, PO11 which had characteristics of humanities. It is necessary that systematic establishment of course completion system in basic design, element design and comprehensive design by giving weight to design education that aims to strengthen design capability should be made in curriculum of engineering education. It was found that there was considerable difference in educational needs about program outcomes especially in PO1(analysis experiment), PO4(problem solving), PO5(practical ability), PO6(teamwork) according to grade, gender and specialty and therefore this should be considered in designing curriculum. It is judged that operation of flexible education program should be arranged if learning achievement through regular curriculum Is limited.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.6
• /
• pp.1081-1090
• /
• 2022

The face recognition model is used for identity recognition of smartphones, providing convenience to many users. As a result, the security review of the DNN model is becoming important, with adversarial attacks present as a well-known vulnerability of the DNN model. Adversarial attacks have evolved to decision-based attack techniques that use only the recognition results of deep learning models to perform attacks. However, existing decision-based attack technique[14] have a problem that requires a large number of queries when generating adversarial examples. In particular, it takes a large number of queries to approximate the gradient. Therefore, in this paper, we propose a method of generating adversarial examples using orthogonal space sampling and dimensionality reduction sampling to avoid wasting queries that are consumed to approximate the gradient of existing decision-based attack technique[14]. Experiments show that our method can reduce the perturbation size of adversarial examples by about 2.4 compared to existing attack technique[14] and increase the attack success rate by 14% compared to existing attack technique[14]. Experimental results demonstrate that the adversarial example generation method proposed in this paper has superior attack performance.

• Journal of Science Education
• /
• v.36 no.2
• /
• pp.235-250
• /
• 2012

This paper aims to find out whether or not elementary school students recognize and understand scientific units that they encounter in their everyday life. To select appropriate units for the survey, first, scientific units in elementary textbooks of science and other science related subjects were analyzed. Then it was examined how these units were related to the learners' daily life. The participants in the current survey were 320 elementary school 6th graders. A questionnaire consisted of 11 units of science, such as kg for mass, km for distance, L for volume, V for voltage, s for time, $^{\circ}C$ for temperature, km/h for speed, kcal for heat, % for percentage, W for electric power, pH for acidity, which can often be seen and used in daily life. The students were asked to do the following four tasks, (1) to see presented pictures and select appropriate scientific units, (2) to write reasons for choosing the units, (3) to answer what the units are used for, and (4) to check where to find the units. The data were analyzed in terms of the percentage of the students who seemed to well recognize and understand the units, using SPSS 17.0 statistical program. The results are as follows: Regarding the general use of the units, it was revealed that almost the same units were repeated in science and other subject textbooks from the same grade. With an increase of the students' grade more difficult units were used. As for the use of each unit, it was found that they seemed to relatively well understand what these units kg, km, L, $^{\circ}C$, kcal, km/h, and W stand for, showing more than 91% right. However, the units of V, s, in particular, %, and pH did not seem to be understood. With respect to the recognition of the units, most students did not recognize such units as L for volume and pH for acidity, probably because the units are difficult at the elementary level in comparison to other scientific units. The students indicated that schools were the best place where they could learn and find scientific units related to life, followed by shops/marts, newspapers/broadcasting, streets/roads, homes, and others in that order. The results show that scientific unit learning should be conducted in a systematic way at school and that teachers can play a major role in improving students' understanding and use of the units.

• Korean Journal of Food Science and Technology
• /
• v.47 no.3
• /
• pp.373-379
• /
• 2015

This study aimed to investigate the anti-amnesic effect of perilla oil against trimethyltin (TMT)-induced learning and memory impairment in ICR mice. Perilla oil (2.5 mL/kg of body weight) and soybean oil (2.5 mL/kg of body weight) were administered orally to mice for 3 weeks, and at the end of the experimental period, cognitive behavior was examined by Y-maze and Morris water maze (MWM) tests. Behavioral tests showed that the mice treated with perilla oil had improved cognitive function compared to that in mice administered soybean oil. Analysis of brain tissue showed that perilla oil significantly lowered acetylcholinesterase activity and malondialdehyde (MDA) levels. Oxidized glutathione (GSH)-to-total GSH ratio also decreased from 10.4% to 5.3% in perilla oil-treated mice, but superoxide dismutase (SOD) activity increased from 11.7 to 14.2 U/mg protein. Therefore, these results suggest that the perilla oil could be a potential functional substance for improving cognitive function.

• Journal of KIISE
• /
• v.44 no.12
• /
• pp.1252-1257
• /
• 2017

In dialogues, speakers' intentions can be represented by sets of an emotion, a speech act, and a predicator. Therefore, dialogue systems should capture and process these implied characteristics of utterances. Many previous studies have considered such determination as independent classification problems, but others have showed them to be associated with each other. In this paper, we propose an integrated model that simultaneously determines emotions, speech acts, and predicators using a convolution neural network. The proposed model consists of a particular abstraction layer, mutually independent informations of these characteristics are abstracted. In the shared abstraction layer, combinations of the independent information is abstracted. During training, errors of emotions, errors of speech acts, and errors of predicators are partially back-propagated through the layers. In the experiments, the proposed integrated model showed better performances (2%p in emotion determination, 11%p in speech act determination, and 3%p in predicator determination) than independent determination models.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.1
• /
• pp.381-385
• /
• 2015

Background:Hepatocellular carcinoma (HCC) is the second most common malignancy in Egypt. Data mining is a method of predictive analysis which can explore tremendous volumes of information to discover hidden patterns and relationships. Our aim here was to develop a non-invasive algorithm for prediction of HCC. Such an algorithm should be economical, reliable, easy to apply and acceptable by domain experts. Methods: This cross-sectional study enrolled 315 patients with hepatitis C virus (HCV) related chronic liver disease (CLD); 135 HCC, 116 cirrhotic patients without HCC and 64 patients with chronic hepatitis C. Using data mining analysis, we constructed a decision tree learning algorithm to predict HCC. Results: The decision tree algorithm was able to predict HCC with recall (sensitivity) of 83.5% and precession (specificity) of 83.3% using only routine data. The correctly classified instances were 259 (82.2%), and the incorrectly classified instances were 56 (17.8%). Out of 29 attributes, serum alpha fetoprotein (AFP), with an optimal cutoff value of ${\geq}50.3ng/ml$ was selected as the best predictor of HCC. To a lesser extent, male sex, presence of cirrhosis, AST>64U/L, and ascites were variables associated with HCC. Conclusion: Data mining analysis allows discovery of hidden patterns and enables the development of models to predict HCC, utilizing routine data as an alternative to CT and liver biopsy. This study has highlighted a new cutoff for AFP (${\geq}50.3ng/ml$). Presence of a score of >2 risk variables (out of 5) can successfully predict HCC with a sensitivity of 96% and specificity of 82%.