• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.159-162
• /
• 2017

Long Short-Term Memory Recurrent Neural Network(LSTM RNN)는 순차 데이터 모델링에 적합한 딥러닝 모델이다. Bidirectional LSTM RNN(BiLSTM RNN)은 RNN의 그래디언트 소멸 문제(vanishing gradient problem)를 해결한 LSTM RNN을 입력 데이터의 양 방향에 적용시킨 것으로 입력 열의 모든 정보를 볼 수 있는 장점이 있어 자연어처리를 비롯한 다양한 분야에서 많이 사용되고 있다. Highway Network는 비선형 변환을 거치지 않은 입력 정보를 히든레이어에서 직접 사용할 수 있게 LSTM 유닛에 게이트를 추가한 딥러닝 모델이다. 본 논문에서는 Highway Network를 한국어 의미역 결정에 적용하여 기존 연구 보다 더 높은 성능을 얻을 수 있음을 보인다.

• Journal of Korean Society of Transportation
• /
• v.6 no.1
• /
• pp.33-41
• /
• 1988

The Wisconsin Department of Transportation currently has a Datalog Information System which facilitates the collection of geometric characteristics at every 0.01 mile high way section. The objectives of this study are to develop the plan and profile drawing of a highway section utilizing the Datalog Information system and to develop a methodology of investigating the safety aspects associated with the highway section being considered. For this purpose, two of the highway design elements, the minimum stoping sight distance as well as passing sight distance based on the AASHTO requirements, were applied in this study. A computer program was also developed to facilitate the data processing activity. The results from the computer program and from a manual analysis which adopted the identical methodology used in the computer program were in a good agreement. a few discrepancies between the two results were due to the data collection error and they were believed to be negligible. Using the computerized methodology developed in this study one does not need the plan and profile drawing to investigate the safety of a highway section, which appears to be an essential progress to the Computer Aided design and Draft in highway engineering field.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• 2008.11a
• /
• pp.922-925
• /
• 2008

Many highway projects are under way in Korea. However, owners frequently find that the project cost exceeds the budget and they are unable to identify the underlining reasons. The main purpose of this research is to develop cost models for transportation projects in Korea using the multiple linear regression (MLR). The data consist of 27 completed transportation projects, built from 1991 to 2001, The technique of multiple regression analysis is used to develop the parametric cost estimating model for total budget cost per highway square meter (TBC/$m^2$). Findings of the study indicated that MLR car be applied to highway projects in Korea. There are twf) major contributions of this research. (1) the identification of transportation parameters as a significant cost driver for transportation costs and (2) the successful development of the parametric cost estimating models for transportation projects in Korea.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.11a
• /
• pp.1140-1143
• /
• 2007

The purpose of this study is the improvement of the prediction model of highway noise. It includes the measurement and analysis of predicted noise levels by various programs in types of road and environments. The results of the measurement are compared with the noise levels predicted by improved highway noise prediction model and domestic prediction models, (Improved highway noise prediction model was considered ASJ-2003, ISO-9613 part2 and noise power of road surface types at Korean highway road.)

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.11a
• /
• pp.101-104
• /
• 2007

All of noise prediction model have it's own features in the case of modeling conditions, so it is very important to know the features of each model case by case for a proper modeling, especially using at the Environmental Impact Assessment. For prediction of highway traffic noise and abating the noise by barriers, two kinds of prediction model, HW-NOISE, KHTN(Korea Highway Traffic Noise) has been mainly used in Korea. In this study, the features of these models were described at the same conditions. The properties of sound power from a road, diffraction characteristics from a barrier, sound pressure level decaying in each model were investigated. Using the results, it will be anticipated that the proper using of prediction models in the works of highway noise abating.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.11a
• /
• pp.30-33
• /
• 2006

The purpose of this study is the improvement of the prediction model of highway noise. It includes the measurement and analysis of the highway noise for various types of roads and environments. The results of the measurement are compared with the noise levels predicted by highway noise prediction models including domestic and foreign ones and any recommendation to improve the accuracy of the models should be provided.

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.159-162
• /
• 2017

Long Short-Term Memory Recurrent Neural Network(LSTM RNN)는 순차 데이터 모델링에 적합한 딥러닝 모델이다. Bidirectional LSTM RNN(BiLSTM RNN)은 RNN의 그래디언트 소멸 문제(vanishing gradient problem)를 해결한 LSTM RNN을 입력 데이터의 양 방향에 적용시킨 것으로 입력 열의 모든 정보를 볼 수 있는 장점이 있어 자연어처리를 비롯한 다양한 분야에서 많이 사용되고 있다. Highway Network는 비선형 변환을 거치지 않은 입력 정보를 히든레이어에서 직접 사용할 수 있게 LSTM 유닛에 게이트를 추가한 딥러닝 모델이다. 본 논문에서는 Highway Network를 한국어 의미역 결정에 적용하여 기존 연구 보다 더 높은 성능을 얻을 수 있음을 보인다.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2005.11a
• /
• pp.355-358
• /
• 2005

Corrosion of reinforced concrete structures in marine environment is one of the most important mechanisms of deterioration. Under Korean highway bridge, the time for the steel reinforcement in the concrete to exhibit initial signs of corrosion is within three decades. Therefore, 'SUM JIN' highway bridge, located in a corrosive marine environment on the south of Korea, had been examined the current condition of the steel reinforcement corrosion in concrete by half-cell potentials, chloride contamination of concrete and so on. According to the tests, the protecting film around the reinforcement is deteriorated and corrosion activity developed in tidal zone. The purpose of this paper is to report the effects of 'SUM JIN' highway bridge damaged by chlodide attack and to present the results of repair of 'SUM JIN' highway concrete bridge in domestic marine environment.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2006.05a
• /
• pp.626-629
• /
• 2006

Suk-San highway bridge, located on a soft ground environment, had been examined the current condition of settlement estimation throughout G.P.R(Ground Penetrating Radar), general observation and visual observation(video camera & scope). According to the above observations, the ground of this area has sunk about thirty centimeters since 1996. Also, currently, Suk-San highway bridge has been disjoining the gap between the structure and ground. Therefore, it is necessary to fill it up the gap. The purpose of this paper is to report the effects of Sunk-San highway bridge was observed by G.P.R. & general observation etc. and to present the results of repair of Suk-San highway bridge filling the gap up.

• Proceedings of the KIEE Conference
• /
• 1999.07b
• /
• pp.680-682
• /
• 1999

In this paper, we propose an traffic data acquisition method and automatic vehicle classification system using laser. We use a invisible laser to minimize measuring error caused by environmental variation. also we use radio frequency data communication and PCMCIA for operating convenience.