Papers by Naohiko Kohtake
2017 6th IIAI International Congress on Advanced Applied Informatics (IIAI-AAI)
This paper describes a method of predicting the rutting depth by introducing Adam and dropout int... more This paper describes a method of predicting the rutting depth by introducing Adam and dropout into MLP of neural network and GRU of recurrent neural net that can handle time series data. We built a model to predict the current rutting depth from the past rutting 3 years ago. We compared RMSE with the multiple regression model (MLR), which is most frequently used as a regression problem for the time variation of rutting depth. As a result, RMSE decreased in the order of MLR, MLP, GRU. The difference between GRU and RMSE of MLR was about 10%.
2021 Second International Symposium on Instrumentation, Control, Artificial Intelligence, and Robotics (ICA-SYMP)
The control of economic activities by governments for the purpose of minimizing COVID-19 spread i... more The control of economic activities by governments for the purpose of minimizing COVID-19 spread influenced atmospheric conditions. Satellite-based technology is promising to monitor these changes and even non-specialists in geospatial analysis can use it through cloud-based open platforms. This study aims to give scientific interpretation on the causes of the atmospheric changes based on these governmental controls on human activity. The nitrogen dioxide (NO2) levels were monitored in Ayutthaya Province in Thailand in an urban area including industrial zones during January 1 to April 30, 2020. An analysis was conducted with Google Earth Engine by using four datasets including NO2 data from Sentinel-5P. The mean value of NO2 density in 2020 decreased 12.5% compared with that in 2019. The closure of industrial facilities by the government seemed to be a key cause of the phenomena. The decline was statistically tested and resulted in a significant difference between the two periods. As the new daily cases decreased to single digits, the difference became non-significant from May 1, 2020. Single timeseries analysis on NO2 density change enabled visualization of the brief impact of responses and policies made by the government and subsequent economic activities; however, further atmospheric and industrial activity data are required for further assessment.
2017 Annual IEEE International Systems Conference (SysCon)
This study examines an approach to model and analyze the transforming health-information system o... more This study examines an approach to model and analyze the transforming health-information system of systems. The process and architecture for the Cambodian malaria-surveillance system were modeled using the Engineering Systems Multiple-Domain Matrix modeling framework. By using the attributes of the process, architecture, and risk associated with the environment within the model, relative weights of the constituent systems were scored at each interval of time. The simulated confidence intervals of the absolute difference in the scores indicated that this approach captured the transformation of the system under investigation. This approach provides the first step in analyzing the transitional conditions of constituent systems as well as the entire health-information system, whereby informed decision-making for optimizing continuous system management is facilitated.
International Journal of Advanced Research in Engineering
The pavement deterioration prediction model is a basic module of the PMS (Pavement Management Sys... more The pavement deterioration prediction model is a basic module of the PMS (Pavement Management System), and its prediction results influence the decision making of pavement administrators. Hence, it is very important to improve prediction accuracy. There is a rutting depth as a major indicator showing the state of pavement used in PMS. In this research, we propose a method to predict rutting depth by introducing a technique to improve prediction accuracy by suppressing over-fitting by dropout and gradient clipping in the recently rapidly developing NN (Neural Network) model. In addition, since pavement survey data are time series data that were inspected at the same place multiple times, we applied the RNN (Recurrent Neural Network) model which can model time series data. The proposed method was applied to rutting depth prediction of periodic survey data of Kawasaki city in Japan from 1987 to 2016. In order to compare prediction accuracy, we predicted three years later using proposed...
Abstract. This paper introduces the concept of DIY (do-it-yourself) pervasive services that enabl... more Abstract. This paper introduces the concept of DIY (do-it-yourself) pervasive services that enable non-experts to establish pervasive computing environments for everyday life. The aim of this concept is that non-expert users can prepare the essential hardware and software for pervasive services, assemble them at appropriate positions, and use them practically. As the first approach, this concept has been implemented in the u-Texture II, which is a self-organizable panel created beforehand as a smart building block. The assembled u-Texture IIs create various pervasive computing environments and each environment provides suitable pervasive services without experts. 1
2017 IEEE International Conference on Big Data (Big Data)
Smallholder agriculture is vital for food security and poverty reduction. Poor, unbanked smallhol... more Smallholder agriculture is vital for food security and poverty reduction. Poor, unbanked smallholder farmers face challenges in obtaining loans for agriculture due to missing financial credit history. Although credit decisions tools with alternatives such as big data from mobile phones currently exist, these are mainly targeted towards consumer credit solutions. There is need for research focused on applying such data to credit decisions in agriculture, considering the unique challenges of the field. This research proposes a credit decision tool using alternative data from a mobile phone application used for reporting agricultural activities. 213 users submitted 41613 reports of farm activity by 11336 farmers over a 5-year period. Users were classified groups based on personal information, association with farmers, and reports. Three methods are applied for analysis: multiple logistic regression, a support vector machine with a linear kernel, and a support vector machine with a radial basis function kernel. Three sets of features, selected with different methods, are used. The results are evaluated by accuracy, and Area Under the Receiver operating curve (AUC). The support vector machine with a radial basis function kernel shows the best performance on the original data set with an Area Under the Receiver operating curve (AUC) value of 0.983. Feasibility of applying each method in agricultural microfinance is considered. Finally, further development of the models using Geographic Information System analysis of geospatial data which affects agricultural productivity, such as soil type and water resources, is suggested to incorporate the ability of the borrower to repay loans from farm activity.
2021 Second International Symposium on Instrumentation, Control, Artificial Intelligence, and Robotics (ICA-SYMP), 2021
The control of economic activities by governments for the purpose of minimizing COVID-19 spread i... more The control of economic activities by governments for the purpose of minimizing COVID-19 spread influenced atmospheric conditions. Satellite-based technology is promising to monitor these changes and even non-specialists in geospatial analysis can use it through cloud-based open platforms. This study aims to give scientific interpretation on the causes of the atmospheric changes based on these governmental controls on human activity. The nitrogen dioxide (NO2) levels were monitored in Ayutthaya Province in Thailand in an urban area including industrial zones during January 1 to April 30, 2020. An analysis was conducted with Google Earth Engine by using four datasets including NO2data from Sentinel-5P. The mean value of NO2density in 2020 decreased 12.5% compared with that in 2019. The closure of industrial facilities by the government seemed to be a key cause of the phenomena. The decline was statistically tested and resulted in a significant difference between the two periods. As t...
As one of the JAXA’s software independent verification and validation(IV&V) activities, test anal... more As one of the JAXA’s software independent verification and validation(IV&V) activities, test analysis has been done for dependable flight softwares. The aim of it is to establish the methodology to identify "what to test" efficiently. The analysis is based on "Viewpoint of Testing" which means various aspects or conditions to be tested. This paper introduces our experience in viewpoint-based test analysis of some dependable flight softwares and discuss the results. I. OBJECTIVES OR GOALS OF THE WORK The purpose of this activity is to establish the methodology to identify "what to test" efficiently for spacecraft on-board software. To identify "what to test" is very important because the chance of revealing software faults depends on it. Because of this background, we introduce the concept of "Viewpoint of Testing", which involves various aspects or conditions to be tested; these aspects and conditions include quality characteristics,...
With context-aware computing, a user can control appliances automatically based on his/her contex... more With context-aware computing, a user can control appliances automatically based on his/her context. Automatic control generates conflict of demands. Thus, mutual exclusion is needed for the appliance which cannot fill the demands simultaneously. We show the technique of flexible mutual exclusion based on the relationship among users and the place where the appliance exists.
The pavement deterioration prediction model is a basic module of the PMS (Pavement Management Sys... more The pavement deterioration prediction model is a basic module of the PMS (Pavement Management System), and its prediction results influence the decision making of pavement administrators. Hence, it is very important to improve prediction accuracy. There is a rutting depth as a major indicator showing the state of pavement used in PMS. In this research, we propose a method to predict rutting depth by introducing a technique to improve prediction accuracy by suppressing over-fitting by dropout and gradient clipping in the recently rapidly developing NN (Neural Network) model. In addition, since pavement survey data are time series data that were inspected at the same place multiple times, we applied the RNN (Recurrent Neural Network) model which can model time series data. The proposed method was applied to rutting depth prediction of periodic survey data of Kawasaki city in Japan from 1987 to 2016. In order to compare prediction accuracy, we predicted three years later using proposed...
The Graduate School of System Design and Management of Keio University (Keio SDM) was established... more The Graduate School of System Design and Management of Keio University (Keio SDM) was established on April 1, 2008 to cultivate systemic thinkers who can lead in the development and operation of large-scale complex technological and social systems. One of the indispensable capabilities of system thinker is multi viewpoints. No system can be described from only one viewpoint. Complex technological system and social system have many stakeholders. To understand and satisfy stakeholder's requirements, the system thinkers have to have multi-viewpoints. Our approach to cultivate persons who have multi-viewpoints through a class at KEIO SDM is to make good use of combination of individual work and group work. Our students in the Keio SDM are of all ages and come from a wide variety of backgrounds, including employees and officials (both young and experienced) from private public sectors. We also have recent graduates. Group works with different background can cultivate multi viewpoints...
Abstract—Retirement from sport can be stressful to athletes for many reasons. Accordingly, it is ... more Abstract—Retirement from sport can be stressful to athletes for many reasons. Accordingly, it is necessary to match coping strategies depending on the stressors. One of the athlete career assistance programs for Japanese top athletes in Japan, the Japan Olympic Committee Career Academy (JCA), has focused on the service contents regarding occupational supports which can be said to cope with financial and occupational stress; however, other supports such as psychological support were unclear due to the lack of psychological professionals in the JCA. Tailoring the program, it is important to match the needs of the athletes at athletic retirement with the service contents. Japanese Olympic athletes have been found to retire for different reasons. Especially female athletes who competed in the Summer Olympic Games were found to retire with psychological reasons. The purpose of this research was to investigate the types of stressors Japanese female athletes experience as a result of athle...
2018 IEEE International Systems Engineering Symposium (ISSE), 2018
Despite the substantial efforts of various stakeholders, numerous people do not have sufficient a... more Despite the substantial efforts of various stakeholders, numerous people do not have sufficient access to healthcare services. Though the reliability and timeliness of the health information collected in the healthcare system play critical roles in realizing appropriate healthcare resource allocation, the quality of the health information continues to be of concern. Because the health information system is a system-of-systems that transforms over time in response to the changing environment and needs of the stakeholders, it is important to understand the conditions of transitioning it continuously to improve the data quality. This study examined the continuous management of the health information system-of-systems by the engineering systems multiple-domain modeling approach. The engineering systems multiple-domain matrix framework was applied to model the Cambodian malaria surveillance system. By including the spatiotemporal dynamics of the environmental attributes in the model alon...
An apparatus for controlling a plurality of injection molding machines is disclosed wherein each ... more An apparatus for controlling a plurality of injection molding machines is disclosed wherein each injection molding machine is provided with a digital memory. The memories are provided with data from a central unit and are activated as a function of travel of appropriate machine parts of the respective molding machines.
This paper proposes a new methodology to design a public policy for re-vitalizing declined rural ... more This paper proposes a new methodology to design a public policy for re-vitalizing declined rural towns by using a systems-approach to visualize and quantify the residents" preferences on commu- nity-building. The new methodology, named the Interactive Social Design Model (ISDM), roots in the Vee model (Forsberg et al.(1)), a standard framework to describe the life-cycle of systems developments, and Ostrom (2)"s commons theory. The ISDM provides a new perspective for the re- gional policy since enhanced social capitals with stimulated social interactions are mobilized as the real engines for regional economic growth and em- ployments. This paper selected the Akita City, a core city in the north-eastern region of Japan now in decay, as the social research field to validate this new methodology. The collected data there and our analysis proved well that the ISDM is effective to accomplish the policy goal for regional re-vitalization.
Credit scoring of financially excluded persons is challenging for financial institutions because ... more Credit scoring of financially excluded persons is challenging for financial institutions because of a lack of financial data and long physical distances, which hamper data collection. The remote collection of alternative data has the potential to overcome these challenges, enabling credit access for such individuals. Whereas alternative data sources such as mobile phones have been investigated by previous researchers, this research proposes the integration of mobile-phone, satellite, and public geospatial data to improve credit evaluations where financial data are lacking. An approach to integrating these disparate data sources involving both spatial and temporal analysis methods such as spatial aggregation was employed, resulting in various data combinations. The resulting data sets were used to train classifiers of varying complexity, from logistic regression to ensemble learning. Comparisons were based on various performance metrics, including accuracy and the area under the rece...
Area Business Continuity Management integrates both private and public sectors under the concept ... more Area Business Continuity Management integrates both private and public sectors under the concept of Public-Private Partnership against common risks including natural disasters while previous research mainly focused within private sectors or critical infrastructures. Its issue is that the broader range of heterogeneous stakeholders gets involved the more complex and ambiguous their connectivity becomes. Although, a bottom up approach has been proposed that identifies interdependency of stakeholders by aggregating their dependent activities on business continuities, a practical method has not been proposed yet. This research therefore designed a tangible method which enables to elicit dependencies of individual stakeholder by extending ideas of Business Impact Analysis and Risk Assessment. The proposed method was verified through a workshop conducted in an educational program of disaster risk management. It was observed that the method enables to identify not only external dependent a...
Uploads
Papers by Naohiko Kohtake