TELKOMNIKA Telecommunication Computing Electronics and Control Vol. 20, No. 6, December 2022, pp. 1276~1287 ISSN: 1693-6930, DOI: 10.12928/TELKOMNIKA.v20i6.24087  1276 Journal homepage: http://telkomnika.uad.ac.id A proposal model using deep learning model integrated with knowledge graph for monitoring human behavior in forest protection Van Hai Pham 1 , Quoc Hung Nguyen 2 , Thanh Trung Le 2 , Thi Xuan Dao Nguyen 2 , Thi Thuy Kieu Phan 2 1 Faculty of Computer Science, School of Information and Communication Technology, Hanoi University of Science and Technology, Hanoi, Vietnam 2 School of Business Information Technology, University of Economics Ho Chi Minh City (UEH), Ho Chi Minh City, Vietnam Article Info ABSTRACT Article history: Received Jul 20, 2021 Revised Sep 14, 2022 Accepted Sep 24, 2022 In conventional monitoring of human behavior in forest protection, deep learning approaches can be detected human behavior significantly since thousands of visitors’ forest protection is abnormal and normal behaviors coming to national or rural forests. This paper has presented a new approach using a deep learning model integrated with a knowledge graph for the surveillance monitoring system to be activated to confirm human behavior in a real-time video together with its tracking human profile. To confirm the proposed model, the proposed model has been tested with data sets through case studies with real-time video of a forest. The proposed model provides a novel approach using face recognition with its behavioral surveillance of the human profile integrated with the knowledge graph. Experimental results show that the proposed model has demonstrated the model’s effectiveness. Keywords: Deep learning Forest protection Human action recognition Identifying human behavior Video time-lapse This is an open access article under the CC BY-SA license. Corresponding Author: Van Hai Pham Faculty of Computer Science, School of Information and Communication Technology Hanoi University of Science and Technology, Hanoi, Vietnam Email: haipv@soict.hust.edu.vn 1. INTRODUCTION Recently, it is hard for a human to protect forests, indiscriminately cutting down makes forest resources recover and become more and more exhausted, many places where forests can no longer regenerate, the land becomes more reclaimed. The role of forests in environmental forest protection is significant to the world. Identifying human behavior using advanced technology has become an important area of research to create or improve applications that monitor human activity. The behavioral human is a time series of graphs, which is significant for long-term monitoring results. This knowledge graph can be kept track of human actions for human behaviors. Knowledge graphs that represent structural relations among entities such as places, actions, geography nodes, and other attributes of human profiles. In addition, a knowledge graph has represented an object such as entities, relationships, and semantic descriptions. Computer vision using either analysis or machine learning approaches is automatically detected face and human behavior in real-time. Knowledge graphs represent object types consisting of (i.e. places, geometrics, image coordinates, and locations) and are considered as Neo4j software for making the graphs. After identifying attributes and objects in the graph, the relationships between objects by using geometric and graph attributes. There have been many studies suggesting some solutions to detect bad behavior of humans who may destroy the forest in a forest domain. In this paper, we have proposed a novel approach using a deep learning model integrated with a knowledge graph for the surveillance monitoring system to be activated to confirm human behavior in a real-time video together with its tracking human profile.