• Journal of Bio-Environment Control
• /
• v.31 no.1
• /
• pp.1-7
• /
• 2022

Strawberry is a stand-out cultivating fruit in Korea. The optimum production of strawberry is highly dependent on growing environment. Smart farm technology, and automatic monitoring and control system maintain a favorable environment for strawberry growth in greenhouses, as well as play an important role to improve production. Moreover, physiological parameters of strawberry plant and it is surrounding environment may allow to give an idea on production of strawberry. Therefore, this study intends to build a machine learning model to predict strawberry's yield, cultivated in greenhouse. The environmental parameter like as temperature, humidity and CO₂ and physiological parameters such as length of leaves, number of flowers and fruits and chlorophyll content of 'Seolhyang' (widely growing strawberry cultivar in Korea) were collected from three strawberry greenhouses located in Sacheon of Gyeongsangnam-do during the period of 2019-2020. A predictive model, Lasso regression was designed and validated through 5-fold cross-validation. The current study found that performance of the Lasso regression model is good to predict the number of flowers and fruits, when the MAPE value are 0.511 and 0.488, respectively during the model validation. Overall, the present study demonstrates that using AI based regression model may be convenient for farms and agricultural companies to predict yield of crops with fewer input attributes.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.17-20
• /
• 2021

본 연구는 인원 계수에 딥러닝 알고리즘을 적용한다. 인원 계수는 안전 관리 분야, 상업 분야에 적용될 수 있다. 예를 들어, 건물 내 화재 발생 시, 계수된 인원을 활용하여 인명 피해를 최소화할 수 있다. 다른 예로, 유동인구 데이터를 기반으로 상권을 분석하여 경제적 효율성을 극대화할 수 있다. 이처럼 인원 데이터의 중요성이 증가함에 따라 인원 계수 연구도 활발하다. 그 예로, 객체 탐지(Object Detection) 같은 딥러닝 기반 인원 계수, 센서 기반 인원 계수 등이 있다. 본 연구에선 딥러닝 알고리즘인 VGGNet을 사용하여 인원을 계수했다. 결과로 Mean Absolute Percentage Error(이하 MAPE)는 약 5.9%의 오차율을 보였다. 결과 확인 방법으로는 설명 가능한 인공지능(XAI) 알고리즘 중 하나인 Grad-CAM을 적용했다.

• Journal of Digital Convergence
• /
• v.16 no.5
• /
• pp.137-143
• /
• 2018

Korea's air passenger traffic has been growing steadily. In this paper, we propose a forecasting model of air passenger demand to ascertain the growth trend of air passenger transportation performance in Korea. We conducted a simulation based on System Dynamics with the demand as a dependent variable, and international oil prices, GDP and exchange rates as exogenous variables. The accuracy of the model was verified using MAPE and $R^2$, and the proposed prediction model was verified as an accurate prediction model. As a result of the demand forecast, it is predicted that the air passenger demand in Korea will continue to grow, and the share of low cost carriers will increase sharply. The addition of the Korean transportation performance of foreign carriers in Korea and the transportation performance of Korean passengers due to the alliance of airlines will provide a more accurate forecast of passenger demand.

• Journal of Korea Water Resources Association
• /
• v.54 no.12
• /
• pp.1275-1284
• /
• 2021

This paper presents a machine learning technique applied to prediction of time-dependent local scour around bridge piers in both non-cohesive and cohesive beds. The support vector machines (SVM), which is known to be free from overfitting, is used. The time-dependent scour depths are expressed by 7 and 9 variables for the non-cohesive and cohesive beds, respectively. The SVM models are trained and validated with time series data from different sources of experiments. Resulting Mean Absolute Percentage Error (MAPE) indicates that the models are trained and validated properly. Comparisons are made with the results from Choi and Choi's formula and Scour Rate in Cohesive Soils (SRICOS) method by Briaud et al., as well as measured data. This study reveals that the SVM is capable of predicting time-dependent local scour in both non-cohesive and cohesive beds under the condition that sufficient data of good quality are provided.

• Journal of Navigation and Port Research
• /
• v.44 no.6
• /
• pp.517-523
• /
• 2020

Pyeongtaek·Dangjin port handles the largest volume of finished vehicles in Korea, including more than 95% of imported cars. However, since the volume of imported cars has been stagnant since 2015, officials planning to invest in port development or automobile-related industries must make new forecasts. Economic variables such as the GDP often have been used in predicting automobile volume, but prior research showed that the impact of these economic variables on automobile volume I has been gradually decreasing in developed countries. These variables remain important predictors, however, in developing countries that experience rapid economic growth. In this study, predicting the volume of imported passenger cars at Pyeongtaek·Dangjin port, the decreasing Korean population was a major factor we considered. Our forecast showed that the volume of imported passenger cars at Pyeongtaek·Dangjin port will gradually decrease -by 2021. The Mean Absolute Percentage Error (MAPE) verification was performed to measure the accuracy of the predicted results, and the scenario analysis was performed on the share of imported passenger cars.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.4
• /
• pp.1-13
• /
• 2023

This study develops an optimal mobile-communication-based National Highway traffic volume estimation model using an ensemble-based machine learning algorithm. Based on information such as mobile communication data and VDS data, the LightGBM model was selected as the optimal model for estimating traffic volume. As a result of evaluating traffic volume estimation performance from 96 points where VDS was installed, MAPE was 8.49 (accuracy 91.51%). On the roads where VDS was not installed, traffic estimation accuracy was 92.6%.

• The Journal of Bigdata
• /
• v.8 no.2
• /
• pp.97-107
• /
• 2023

Small business owners are relatively likely to be alienated from various benefits caused by the change to a big data/AI-based society. To support them, we would like to detect a hot place based on the floating population to support small business owners' decision-making in the start-up area. Through various studies, it is known that the population size of the region has an important effect on the sales of small business owners. In this study, inland regions were extracted from the Incheon floating population data from January 2019 to June 2022. the Data is consisted of a grid of 50m intervals, central coordinates and the population for each grid are presented, made image structure through imputation to maintain spatial information. Spatial outliers were removed and imputated using LOF and GAM, and temporal outliers were removed and imputated through LOESS. We used ConvLSTM which can take both temporal and spatial characteristics into account as a predictive model, and used AutoEncoder structure, which performs outliers detection based on reconstruction error to define an area with high MAPE as a hot place.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.3
• /
• pp.166-175
• /
• 1998

When we make a speaker adapted model using MAPE (maximum a posteriori estimation), the adapted model has one mixture in each state. This is because we cannot estimate a number of a priori distribution from a speaker-independent model in each state. If the model is represented by one mixture in each state, it is not well adadpted to specific speaker because it is difficult to represent various speech informationof the speaker with one mixture. In this paper, we suggest the method using several mixtures to well represent various speech information of the speaker in each state. But, because speaker-specific training dat is not sufficient, this method can't be used in every state. So, we make the number of mixtures in each state variable in proportion to the number of frames and to the determinant ofthe variance matrix in the state. Using the proposed method, we reduced the error rate than methods using one branch in each state.

• Communications for Statistical Applications and Methods
• /
• v.30 no.3
• /
• pp.273-289
• /
• 2023

In this paper, we develop a new time series model for predicting IPO (initial public offering) data with non-negative integer value. The proposed model is based on integer-valued autoregressive (INAR) model with a Poisson thinning operator. Just as the heterogeneous autoregressive (HAR) model with daily, weekly and monthly averages in a form of cascade, the integer-valued heterogeneous autoregressive (INHAR) model is considered to reflect efficiently the long memory. The parameters of the INHAR model are estimated using the conditional least squares estimate and Yule-Walker estimate. Through simulations, bias and standard error are calculated to compare the performance of the estimates. Effects of model fitting to the Korea's IPO are evaluated using performance measures such as mean square error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE) etc. The results show that INHAR model provides better performance than traditional INAR model. The empirical analysis of the Korea's IPO indicates that our proposed model is efficient in forecasting monthly IPO volumes.

• Journal of Urban Science
• /
• v.7 no.1
• /
• pp.45-54
• /
• 2018

The purpose of this study is to find the factors that reduce prediction error in traffic volume using highway traffic volume data. The ARIMA model was used to predict the day, and it was confirmed that weekday and weekly characteristics were distinguished by prediction error. The forecasting results showed that weekday characteristics were prominent on Tuesdays, Wednesdays, and Thursdays, and forecast errors including MAPE and MAE on Sunday were about 15% points and about 10 points higher than weekday characteristics. Also, on Friday, the forecast error was high on weekdays, similar to Sunday's forecast error, unlike Tuesday, Wednesday, and Thursday, which had weekday characteristics. Therefore, when forecasting the time series belonging to Friday, it should be regarded as a weekly characteristic having characteristics similar to weekend rather than considering as weekday.