http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : 황하은(Ha Eun Hwang) (검색결과 1 건)
- 무료
- 기관 내 무료
- 유료
-
김서연(Seo-Yeon Kim),이민정(Minjung Lee),이지윤(Jiyoon Lee),황하은(Ha Eun Hwang),장광호(Gwang Ho Jang),김희두(Hee Dou Kim),김종윤(Jong Yoon Kim),김성범(Seoung Bum Kim) 대한산업공학회 2021 대한산업공학회지 Vol.47 No.5
Maintaining a safe society is a fundamental value of the nation, and is the primary purpose of policing. Recently, a new approach of using artificial intelligence for policing, known as smart policing, has been proposed. While various studies on smart policing have been discussed, most studies only focus on predicting the hotspot regions or the number of major crimes. The Republic of Korea has a policing system called the 112 system, and the need to incorporate smart policing to the 112 system is increasing gradually. However, the current 112 system is operating less efficiently by deploying the same number of police forces regardless of the number of expected 112 emergency calls. Moreover, studies related to the number of 112 emergency calls are not only insufficient, but also have limitations in that the number of emergency calls in the past has not been utilized. In this study, we propose a multimodal stacking model (MSM) that can predict 112 emergency calls by reflecting the temporal characteristics of the policing data and facilitate interpretation of the important variables. A deep learning-based recurrent neural network and a gradient boosting model are used together in MSM. The proposed MSM yielded excellent predictive performance in experiments using various kinds of regional characteristic data and 112 emergency call data.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
