• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.4
• /
• pp.163-170
• /
• 2019

Korea language has postpositions such as eul, reul, yi, ga, wa, and gwa, which are attached to nouns and add meaning to the sentence. When foreign notations or abbreviations are included in sentences, the appropriate postposition for the pronunciation of the foreign words may not be used. Sometimes, for natural expression of the sentence, two postpositions are used with one in parentheses as in "eul(reul)" so that both postpositions can be acceptable. This study finds examples of using unnatural postpositions when foreign words are included in Korean sentences and proposes a method for using natural postpositions by learning the final consonant pronunciation of nouns. The proposed method uses a recurrent neural network model to naturally express postpositions connected to foreign words. Furthermore, the proposed method is proven by learning and testing with the proposed method. It will be useful for composing perfect sentences for machine translation by using natural postpositions for English abbreviations or new foreign words included in Korean sentences in the future.

• Journal of the Korean Data Analysis Society
• /
• v.20 no.6
• /
• pp.2865-2872
• /
• 2018

In the field of data mining technique, there are various methods such as association rules, cluster analysis, decision tree, neural network. Among them, association rules are defined by using various association evaluation criteria such as support, confidence, and lift. Agrawal et al. (1993) first proposed this association rule, and since then research has been conducted by many scholars. Recently, studies related to crossover entropy have been published (Park, 2016b). In this paper, we proposed a purely symmetric J measure considering directionality and purity in the previously published J measure, and examined its usefulness by using examples. As a result, it is found that the pure symmetric J measure changes more clearly than the conventional J measure, the symmetric J measure, and the pure crossover entropy measure as the frequency of coincidence increases. The variation of the pure symmetric J measure was also larger depending on the magnitude of the inconsistency, and the presence or absence of the association was more clearly understood.

• Journal of The Korean Association of Information Education
• /
• v.25 no.5
• /
• pp.813-824
• /
• 2021

To improve educational performance by analyzing the learning trends of adult learners of Open High Schools, various algorithm models using artificial intelligence were designed and performance was evaluated by applying them to real data. We analyzed Log data of 115 adult learners in the cyber education system of Open High Schools. Most adult learners of Open High Schools learned more than recommended learning time, but at the end of the semester, the actual learning time was significantly reduced compared to the recommended learning time. In the second half of learning, the participation rate of VODs, formation assessments, and learning activities also decreased. Therefore, in order to improve educational performance, learning time should be supported to continue in the second half. In the latter half, we developed an artificial intelligence algorithm models using Tensorflow to predict learning time by data they started taking the course. As a result, when using CNN(Convolutional Neural Network) model to predict single or multiple outputs, the mean-absolute-error is lowest compared to other models.

• Journal of Ocean Engineering and Technology
• /
• v.35 no.2
• /
• pp.141-149
• /
• 2021

The A60 class deck penetration piece is a fire-resistance apparatus installed on the deck compartment to protect lives and to prevent flame diffusion in the case of a fire accident in a ship or offshore plant. In this study, the sensitivity of the fire-resistance performance and approximation characteristics for the A60 class penetration piece was evaluated by conducting a transient heat-transfer analysis and fire test. The transient heat-transfer analysis was conducted to evaluate the fire-resistance design of the A60 class deck penetration piece, and the analysis results were verified via the fire test. The penetration-piece length, diameter, material type, and insulation density were used as the design factors (DFs), and the output responses were the weight, temperature, cost, and productivity. The quantitative effects of each DF on the output responses were evaluated using the design-of-experiments method. Additionally, an optimum design case was identified to minimize the weight of the A60 class deck penetration piece while satisfying the allowable limits of the output responses. According to the design-of-experiments results, various approximate models, e.g., a Kriging model, the response surface method, and a radial basis function-based neural network (RBFN), were generated. The design-of-experiments results were verified by the approximation results. It was concluded that among the approximate models, the RBFN was able to explore the design space of the A60 class deck penetration piece with the highest accuracy.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.20-20
• /
• 2017

최근 기후변화 영향에 따라 전 세계적으로 인명피해 및 재산피해를 유발하는 자연재난이 지속적으로 증가하고 있으며, 그로 인한 자연재해의 규모가 점점 더 커지고 있다. 실제로 우리나라에서도 지난 1994 년에서 2013 년까지 지난 20 년간 자연재해에 의한 피해액은 12조 3천억 원으로 집계되었으며, 이 중 강우와 태풍에 의한 피해가 85 % 이고, 대설에 의한 피해는 약 13 % 로 자연재해 중 대부분의 피해는 강우 및 태풍에서 발생하지만, 폭설에 의한 피해도 적지 않은 것으로 나타났다. 이에 따라, 정확한 예측을 위해 신뢰도 높은 자료 구축을 통한 대설피해 예측에 관한 연구가 필요한 시점이다. 본 연구에서는 대설피해액 예측을 위해 우리나라의 63개 기상 관측소에서 관측한 적설심 자료 및 기상관측 자료와 사회 경제 자료 총 11개를 대설피해 예측을 위한 입력변수로 선정하고, 이를 기상관측소가 속한 도시의 면적에 따라 3개의 지역으로 구분하였다. 주성분분석을 활용하여 선정된 입력변수들을 4개의 주성분으로 구분하고, 인공신경망 및 다중선형 회귀 모형을 구성하여 각 지역별 대설피해 예측의 오차를 분석하였다. 적용결과, 인공신경망 모형을 이용한 대설피해 예측의 수정결정계수는 22.8 %~48.2 %를 나타냈고, 다중선형회귀 모형의 수정결정 계수는 9.2 %~39.7% 로 나타났다. 그러므로 인공신경망 모형이 다중회귀 모형보다 선택된 입력자료를 활용하여 대설피해를 예측하는 목적으로 조금 더 우수한 결과를 나타내었다. 향후 자료를 보완 및 모형의 고도화를 통해 보다 정확한 대설피해 예측 함수 개발이 가능할 것으로 기대된다.

• Korean Journal of Remote Sensing
• /
• v.37 no.1
• /
• pp.85-95
• /
• 2021

Recently, as data mining and artificial neural network techniques are developed, analyzing large amounts of data is proposed to reduce the dimension of the data. In general, empirical orthogonal function (EOF) used to reduce the dimension in the ocean data and recently, Self-organizing maps (SOM) algorithm have been investigated to apply to the ocean field. In this study, both algorithms used the monthly Sea level anomaly (SLA) data from 1993 to 2018 around the East Asia Coasts. There was dominated by the influence of the Kuroshio Extension and eddy kinetic energy. It was able to find the maximum amount of variance of EOF modes. SOM algorithm summarized the characteristic of spatial distributions and periods in EOF mode 1 and 2. It was useful to find the change of SLA variable through the movement of nodes. Node 1 and 5 appeared in the early 2000s and the early 2010s when the sea level was high. On the other hand, node 2 and 6 appeared in the late 1990s and the late 2000s, when the sea level was relatively low. Therefore, it is considered that the application of the SOM algorithm around the East Asia Coasts is well distinguished. In addition, SOM results processed by SLA data, it is able to apply the other climate data to explain more clearly SLA variation mechanisms.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.93-98
• /
• 2021

Due to the development of deep learning and AI, the scale of the model has grown, and it has been integrated into other fields to blend into our lives. However, in environments with limited resources such as embedded devices, it is exist difficult to apply the model and problems such as power shortages. To solve this, lightweight methods such as clouding or offloading technologies, reducing the number of parameters in the model, or optimising calculations are proposed. In this paper, quantization of learned models is applied to ONNX models used in various framework interchange formats, neural network structure and inference performance are compared with existing models, and various module methods for quantization are analyzed. Experiments show that the size of weight parameter is compressed and the inference time is more optimized than before compared to the original model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.8
• /
• pp.993-998
• /
• 2020

The technology of interiors is currently developing around the world. According to various studies, the use of plants to create an environment in the home interior is increasing. However, households using furniture are designed as environment-friendly environment interiors, and in Korea and abroad, plants are used for home interiors. Unexpected accidents are occurring. As a result, there were books and broadcasts about the dangers of specific plants, but until now, accidents continue to occur because they do not properly recognize the dangers of specific plants. Therefore, in this paper, we propose a toxic plant identification system based on a multiplicative neural network model that identifies common toxic plants commonly found in Korea. We propose a high efficiency model. Through this, toxic plants can be identified with higher accuracy and safety accidents caused by toxic plants.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.3
• /
• pp.384-390
• /
• 2020

The reverberation effect on the sound when producing movies or VR contents is a very important factor in the realism and liveliness. The reverberation time depending the space is recommended in a standard called RT60(Reverberation Time 60 dB). In this paper, we propose a scene recognition technique for automatic reverberation editing. To this end, we devised a classification model that independently trains color images and predicted depth images in the same model. Indoor scene classification is limited only by training color information because of the similarity of internal structure. Deep learning based depth information extraction technology is used to use spatial depth information. Based on RT60, 10 scene classes were constructed and model training and evaluation were conducted. Finally, the proposed SCR + DNet (Scene Classification for Reverb + Depth Net) classifier achieves higher performance than conventional CNN classifiers with 92.4% accuracy.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.1
• /
• pp.56-62
• /
• 2021

High forecast accuracy is required as social issues on particulate matter increase. Therefore, many attempts are being made using machine learning to increase the accuracy of particulate matter prediction. However, due to problems with the distribution of imbalance in the concentration and various characteristics of particulate matter, the learning of prediction models is not well done. In this paper, to solve these problems, a binary classification model was proposed to predict the concentration of particulate matter needed for prediction by dividing it into two classes based on the value of 80㎍/㎥. Four classification algorithms were utilized for the binary classification of PM10. Classification algorithms used logistic regression, decision tree, SVM, and MLP. As a result of performance evaluation through confusion matrix, the MLP model showed the highest binary classification performance with 89.98% accuracy among the four models.