Research and develop a data processing system of cow behavioral information based on mounted accelerometers on livestock to monitor and support the assessment of cow health.
-    Focusing on studying the behavior classification algorithm based on acceleration data.
-    Developing, testing the system.

-    Theoretical results:
+ We proposed a Random Forest algorithm with a set of features (root mean square, standard deviation, mean) to classify cow behavior based on synchronized leg and neck acceleration data
+  Scientific report on research on cow behavioral classification based on acceleration data
+  02 articles published in Proceedings of the National Conference
+  01 article published in National Journal
+  01 article published in ISI (ESCI) Journal
+  02 articles published in SCIE Journal
+  02 masters and 02 PhD students in the field of the project.
-    Applied results:
+ A test program was developed to classify cow behavior. The program was tested on data collected, aggregated, and preprocessed from cows at Ba Vi farm (Hanoi, Vietnam).

Critical contribution of this project is combining low-cost components such as leg-mounted and collar-mounted accelerometers to obtain excellent performance for the classification of cow behaviors. Significantly, the acceleration data from these two sensors are synchronized. Therefore, we can substantially expand the amount of information for classification purposes. From our acceleration data, when we applied the Random Forest algorithm with mean, SD, and RMS as features: the performance is excellent. The classification of grazing cow behaviors based on our combining sensors system provides significant improvement from other traditional systems (using only one sensor, leg-mounted or collar-mounted). Our system's development will be an essential part of an IoT-based system that monitors and classifies cow behaviors with high performance.

-    Scientific papers in referred journals (list):
+ ISI Journal (SCI/SCIE):
Phung Cong Phi Khanh, Duc-Tan Tran, Van Tu Duong, Nguyen Hong Thinh, Duc-Nghia Tran*, “The new design of cows' behavior classifier based on acceleration data and proposed feature set”, Mathematical Biosciences and Engineering, Vol. 17, No. 4, March 2020, pp. 2760-2780. ISSN (printed): 1547-1063. ISSN (electronic):1551-0018. DOI: 10.3934/mbe.2020151. (SCIE, Q2)
D. N. Tran, T. N. Nguyen, P. C. P. Khanh and D. T. Tran, "An IoT-based Design Using Accelerometers in Animal Behavior Recognition Systems," in IEEE Sensors Journal, 2021, doi: 10.1109/JSEN.2021.3051194. (SCIE, Q1).             
+ ISI Journal (ESCI):
T. Duc-Nghia, N. Thi-Thu, T. Duc-Tan, "Design of Real-Time Cow Behavior Monitoring System Based on Wireless Sensor Networks and K-Means Clustering Algorithm," Revista Ingeniería Solidaria, ISSN: 2357-6014, vol. 17, no. 3, 2021. doi: https://doi.org/10.16925/2357-6014.2021.03.11 (ISI, ESCI)
+ National Journal:     
Trần Đức Nghĩa*, Đỗ Viết Mạnh, Trần Đức Tân, Nguyễn Cảnh Minh, Phùng Công Phi Khanh, "Phân loại hành vi của động vật dùng thuật toán học máy: tác động của độ dài cửa sổ chồng lấp", Tạp chí Nghiên cứu KH&CN quân sự, Số Đặc san Hội thảo Quốc gia FEE, 10 - 2021, pp. 13-17.
+ National Conference:
Trần Đức Nghĩa*, Nguyễn Việt Anh, Đỗ Viết Mạnh, Bùi Trung Ninh, Trần Văn Thùy, "Nhận dạng hoạt động cho người dựa trên việc sử dụng cảm biến gia tốc ba chiều", Hội thảo quốc gia lần thứ XXIII: Một số vấn đề chọn lọc của Công nghệ thông tin và truyền thông, Nov. 2020, pp. 324-328. ISBN: 978-604-67-1744-7.
T. H. Đào, M. H. Lê, Đ. N. Trần, and Đ. T. Trần, “Xây dựng mạng giám sát hành vi người trong tòa nhà sử dụng công nghệ WIFI,” in REV-ECIT2021, 2021, pp. 48–53.    

-    Technological products (describe in details: technical characteristics, place):
-    Report on Solutions and integrated models of sensors, machine systems serving the recording and processing of cow behavior data
-    Cow behavior classification software (test program)
-    Scientific articles: 2 SCIE and 4 other publications (evidenced in the report summarizing the results of the project)
-    Support in training 02 PhD students and 02 Masters (proven in the summary report on the results of the project)

Proposing pilot implementation at Khanh Phat Agricultural, Livestock and Commercial Cooperative (Ba Vi district, Hanoi city)