• Korean Journal of Agricultural Science
• /
• v.49 no.2
• /
• pp.193-202
• /
• 2022

The recurrent neural network (RNN) algorithm has been widely used in water-related research areas, such as water level predictions and water quality predictions, due to its excellent time series learning capabilities. However, studies on water quality predictions using RNN algorithms are limited because of the scarcity of water quality data. Therefore, most previous studies related to water quality predictions were based on monthly predictions. In this study, the quality of the water in a reservoir in Nonsan, Chungcheongnam-do Republic of Korea was predicted using the RNN-LSTM algorithm. The study was conducted after constructing data that could then be, linearly interpolated as daily data. In this study, we attempt to predict the water quality on the 7th, 15th, 30th, 45th and 60th days instead of making daily predictions of water quality factors. For daily predictions, linear interpolated daily water quality data and daily weather data (rainfall, average temperature, and average wind speed) were used. The results of predicting water quality concentrations (chemical oxygen demand [COD], dissolved oxygen [DO], suspended solid [SS], total nitrogen [T-N], total phosphorus [TP]) through the LSTM algorithm indicated that the predictive value was high on the 7th and 15th days. In the 30th day predictions, the COD and DO items showed R² that exceeded 0.6 at all points, whereas the SS, T-N, and T-P items showed differences depending on the factor being assessed. In the 45th day predictions, it was found that the accuracy of all water quality predictions except for the DO item was sharply lowered.

• Journal of Oriental Neuropsychiatry
• /
• v.19 no.1
• /
• pp.125-146
• /
• 2008

Objectives : The purpose of this study is to suggest for the following experimental study of dementia by reviewing recent oriental medicine journals that have been published since 2000. Methods: We have investigated various types of studies in relation to dementia through 90 articles that have been published from 2000 to 2007 in recent oriental medicine journals were registered Korea research foundation. Results and Conclusions : 1. Since 2000, 88 articles in relation to dementia have been published and almost of them are herbal medicine-centered studies. Also they show a tendency to increase every year. The journal of oriental neuropsychiatry carries the highest number of studies in relation to dementia. 2. According to the experimental paper, there are 30 cases of using herb simplexes, 48 cases of herb-combined prescription, and 10 cases of other ways. Especially 7 cases of using herb-combined prescription relation to Sasang constitution are all for the Taeumin. 3. There are 85 cases of Animal and cellular experimental, 60 cases of using pathologic model induced cytotoxic activity, a case of using L-NAME, 3 cases of 192 saporin, 4 cases of ibotenic acid, 10 cases of focal cerebral ischemia, 3 cases of alcohol-administered, and one case of natural degradation. 4. Moms water maze, Radial arm maze Passive avoidance learning model were using for examining learning and memory of model animal 5. We propose that following studies of dementia are to he investigated of the applied method of using siRNA with tranceduced gene, sample preparation by water-soaking, oriental medical diagnosis, standardization of differentiating symptom and herb simplexes, building the database by classified prescriptions, and experiment model which are based on precise examining mechanism with cell line as like mouse H19-7 hippocampus, rat HT22 hippocampus, astrocyte, microglia, using the model of animals at APP, PS1, BACE, CT99/PS1, APOE4, Tau, APP/PSI/Tau

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.5
• /
• pp.592-601
• /
• 2016

The recurrent neural network (RNN) is a deep learning model which is suitable to sequential or length-variable data. The Long Short-Term Memory (LSTM) mitigates the vanishing gradient problem of RNNs so that LSTM can maintain the long-term dependency among the constituents of the given input sequence. In this paper, we propose a LSTM based language model which can predict following words of a given incomplete sentence to generate a complete sentence. To evaluate our method, we trained our model using multiple Korean corpora then generated the incomplete part of Korean sentences. The result shows that our language model was able to generate the fluent Korean sentences. We also show that the word based model generated better sentences compared to the other settings.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.2
• /
• pp.53-60
• /
• 2012

Nearest-neighbor classification predicts the class of an input data with the most frequent class among the near training data of the input data. Even though nearest-neighbor classification doesn't have a training stage, all of the training data are necessary in a predictive stage and the generalization performance depends on the quality of training data. Therefore, as the training data size increase, a nearest-neighbor classification requires the large amount of memory and the large computation time in prediction. In this paper, we propose a prototype selection algorithm that predicts the class of test data with the new set of prototypes which are near-boundary training data. Based on Tomek links and distance metric, the proposed algorithm selects boundary data and decides whether the selected data is added to the set of prototypes by considering classes and distance relationships. In the experiments, the number of prototypes is much smaller than the size of original training data and we takes advantages of storage reduction and fast prediction in a nearest-neighbor classification.

• Journal of IKEEE
• /
• v.26 no.4
• /
• pp.614-621
• /
• 2022

This paper proposes RIDS (Random Forest-Based Intrusion Detection), which is an intrusion detection system to detect hacking attack based on random forest. RIDS detects three typical attacks i.e. DoS (Denial of service) attack, fuzzing attack, and spoofing attack. It detects hacking attack based on four parameters, i.e. time interval between data frames, its deviation, Hamming distance between payloads, and its diviation. RIDS was designed in memory-centric architecture and node information is stored in memories. It was designed in scalable architecture where DoS attack, fuzzing attack, and spoofing attack can be all detected by adjusting number and depth of trees. Simulation results show that RIDS has 0.9835 accuracy and 0.9545 F1 score and it can detect three attack types effectively.

• Korean Journal of Biological Psychiatry
• /
• v.29 no.2
• /
• pp.33-39
• /
• 2022

Objectives The early growth response 3 (EGR3) gene located in chromosome 8p21.3 is one of the susceptibility loci in many psychiatric disorders. EGR3 gene plays critical roles in signal transduction in the brain, which is involved in neuronal plasticity, neuronal development, learning, memory, and circadian rhythms. Recent studies have suggested EGR3 as a potential susceptibility gene for bipolar disorder (BPD). However, this requires further replication with an independent sample set. Methods To investigate the genetic role of EGR3 in Korean patients, we genotyped six single-nucleotide polymorphisms (SNPs) in the chromosome region of EGR3 in 1076 Korean BPD patients and 773 healthy control subjects. Results Among the six examined SNPs of EGR3 (rs17088531, rs1996147, rs3750192, rs35201266, rs7009708, rs1008949), SNP rs35201266, rs7009708, rs1008949 showed a significant association with BPD (p = 0.0041 for rs35201266 and BPD2, p = 0.0074 for rs1008949 and BPD, p = 0.0052 for rs1008949 and BPD1), which withstand multiple testing correction. In addition, the 'G-C-C-C' and 'G-C-G-C' haplotypes of EGR3 were overrepresented in the patients with BPD (p = 0.0055, < 0.0001, respectively) and the 'G-T-G-C' haplotype of EGR3 was underrepresented in patients with BPD (p = 0.0040). Conclusions In summary, our study supports the association of EGR3 with BPD in Korean population sample, and EGR3 could be suggested as a compelling susceptibility gene in BPD.

• Journal of Korean Society on Water Environment
• /
• v.39 no.1
• /
• pp.46-60
• /
• 2023

Climate change causes fluctuations in water quality in the aquatic environment, which can cause changes in water circulation patterns and severe adverse effects on aquatic ecosystems in the future. Therefore, research is needed to predict and respond to water quality changes caused by climate change in advance. In this study, we tried to predict the dissolved oxygen (DO), chlorophyll-a, and turbidity of the Paldang reservoir for about two weeks using long short-term memory (LSTM) and gated recurrent units (GRU), which are deep learning algorithms based on recurrent neural networks. The model was built based on real-time water quality data and meteorological data. The observation period was set from July to September in the summer of 2021 (Period 1) and from March to May in the spring of 2022 (Period 2). We tried to select an algorithm with optimal predictive power for each water quality parameter. In addition, to improve the predictive power of the model, an important variable extraction technique using random forest was used to select only the important variables as input variables. In both Periods 1 and 2, the predictive power after extracting important variables was further improved. Except for DO in Period 2, GRU was selected as the best model in all water quality parameters. This methodology can be useful for preventive water quality management by identifying the variability of water quality in advance and predicting water quality in a short period.

• Journal of the Korean School Mathematics Society
• /
• v.13 no.1
• /
• pp.23-43
• /
• 2010

This study is to investigate how much highschool students, who have learned functional concepts included in the Middle school math curriculum, understand chapters of the function, to analyze the types of errors which they made in solving the mathematical problems and to look for the proper instructional program to prevent or minimize those ones. On the basis of the result of the above examination, it suggests a classification model for teaching-learning methods and teaching material development The result of this study is as follows. First, Students didn't fully understand the fundamental concept of function and they had tendency to approach the mathematical problems relying on their memory. Second, students got accustomed to conventional math problems too much, so they couldn't distinguish new types of mathematical problems from them sometimes and did faulty reasoning in the problem solving process. Finally, it was very common for students to make errors on calculation and to make technical errors in recognizing mathematical symbols in the problem solving process. When students fully understood the mathematical concepts including a definition of function and learned procedural knowledge of them by themselves, they did not repeat the same errors. Also, explaining the functional concept with a graph related to the function did facilitate their understanding,

• Journal of the Korean Society for Library and Information Science
• /
• v.51 no.4
• /
• pp.99-120
• /
• 2017

In order to develop a semiautomatic support system that allows researchers concerned to efficiently analyze the technical trends for the ever-growing industry and market. This paper introduces a couple of Korean sentence generation models that can automatically generate definitional statements as well as descriptions of technical terms and concepts. The proposed models are based on a deep learning model called LSTM (Long Sort-Term Memory) capable of effectively labeling textual sequences by taking into account the contextual relations of each item in the sequences. Our models take technical terms as inputs and can generate a broad range of heterogeneous textual descriptions that explain the concept of the terms. In the experiments using large-scale training collections, we confirmed that more accurate and reasonable sentences can be generated by CHAR-CNN-LSTM model that is a word-based LSTM exploiting character embeddings based on convolutional neural networks (CNN). The results of this study can be a force for developing an extension model that can generate a set of sentences covering the same subjects, and furthermore, we can implement an artificial intelligence model that automatically creates technical literature.

• Korean Journal of Veterinary Research
• /
• v.48 no.3
• /
• pp.251-258
• /
• 2008

The neurotoxicity of C-terminal fragments of amyloid precusor protein (CT) is known to play some roles in Alzheimer's disease progression. In this study, we investigated the effects of the recombinant C-terminal 105 amino acid fragment of amyloid precusor protein (CT105) on cholinergic function using CT105-injected rat. To study the effects of CT105 on septohippocampal pathway, choline acetyltransferase (ChAT) positive neurons were examined in the medial septum and in the diagonal band after an injection of CT105 peptide into the lateral ventricle. Immunohistological analysis revealed that the number of ChAT-immunopositive cells decreased significantly in both medial septum and diagonal band. In addition, CT105 decreased ChAT-immunopositive cells in the hippocampal area, particulary in the dentate gyros. To study the effect of amyloid beta peptide ($A{\beta}$) and CT105 on the cholinergic system, each peptide was injected into the left lateral ventricle, and acetylcholine (ACh) levels were monitored in hippocampus. ACh level in the hippocampal area was reduced to 60% of control level in $A{\beta}$-treated group, and the level was reduced to 15% of control level in CT105-treated group, at one week after the injection. ACh level was further reduced to 35% of control in $A{\beta}$-treated group, whereas the level was slightly increased to 30% of control in CT105-treated group at 4 weeks after the injection. Taken together, the results in the present study suggest that CT105 impairs the septohippocampal pathway by reducing acetylcholine synthesis and release, which results in damage of learning and memory.