http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
기존 유사 서비스 기능 분석을 통한 차별화된 물류 데이터 분석 플랫폼 설계 및 구현
한기준,정훈,나동길 한국지식정보기술학회 2022 한국지식정보기술학회 논문지 Vol.17 No.2
Despite the rapid growth of the non-face-to-face economy, customized marketing information and sales strategy provision services for small and medium-sized sellers are insufficient, highlighting the need to design and establish an advanced logistic data analytic platform is essential. Therefore, this study analyzed whether there are areas where existing functions can be improved or new functions can be added using big data and AI functions through analysis of domestic logistic analytic platforms. As a result, it seemed that existing services needed to establish data-sharing strategies and further machine learning solutions to investigate multiple internal and external data sources. In addition, in order to secure differentiation from existing services, a machine learning algorithm was designed to predict the weight and price of targeted items based on bill of landing data. Throughout the experiment, we observed that while LSTM does not work well for overall data, the linear regression and MLP models showed good overall performance. It confirmed the models could be used to provide the weight and price prediction services for countries and items targeted to export. The models built in this study are dockerized as a public API service in the future so that researchers and developers can use them actively.
한기준,김영균 대한전자공학회 1982 전자공학회지 Vol.19 No.4
본 논문에서는 thresholding기법을 사용하여 주어진 영상을 이진영상(two-valued image)으로 만든 다음 이 이진 영상에 이중 게이트(double gate)를 적용하여 유사 목표물(targer-like objects)을 추출하는 새로운 기법을 제시하였다. 합성된 영상 데이타에 대해 새로운 기법이 잘 적용, 수행됨을 보였다. This paper describes a new algorithm for the extraction of target-like objects. In order to extract target-like objects, a given image is transformed into two-valued image by thresholding, and then a double gate is applied to the two-valued image. With the synthetic-image data, we have shown the good performance of the new algorithm.