On the afternoon of November 15, 2024, at Room 505, E3 Building, the Information Technology Institute, Vietnam National University Hanoi, held the PhD dissertation defensed for PhD candidate Le Hong Lam, with the topic “Research on Developing Feature Extraction Techniques and Optimizing the Random Forest Model in Detecting Human Fall Events Using Smartphones,” in the field of Information System Management. The dissertation was conducted under the supervision of Assoc. Prof. Nguyen Ha Nam. Below are some images from the dissertation defense.
Attending the defense from the Information Technology Institute were Prof. Tran Xuan Tu – Director; Dr. Le Quang Minh – Deputy Director; and Dr. Duong Quang Khanh – Head of the Department of Science, Technology, and Training.
The dissertation evaluation council included Prof. Tran Xuan Tu – Information Technology Institute, Chairman of the Council; Dr. Le Quang Minh – Information Technology Institute, Secretary. The reviewers were Assoc. Prof. Nguyen Viet Anh – Vietnam Academy of Science and Technology; Assoc. Prof. Pham Van Cuong – Academy of Posts and Telecommunications; Assoc. Prof. Pham Ngoc Hung – University of Danang, University of Science and Technology, VNU Hanoi. The members of the council included Assoc. Prof. Tran Cao Truong – Academy of Military Engineering; Assoc. Prof. Nguyen Thanh Tung – International School, VNU Hanoi.
The dissertation has made the following new contributions:
- Proposes a new feature extraction method from three-axis accelerometer data by combining time-domain and frequency-domain features along with orientation and Hjorth parameters.
- Proposes the use of a genetic algorithm (GA) to optimize the parameters for the Random Forest model (the GA4RF algorithm). GA4RF demonstrates the ability to search for optimal hyperparameters thanks to its chromosome structure, search space, and fitness function designed in a new direction suitable for fall detection datasets.
After hearing the PhD candidate’s presentation and answering questions, the council discussed and concluded that the content and form of the dissertation met all the requirements of a PhD dissertation in Information System Management. The council congratulated PhD candidate Le Hong Lam for successfully defending the PhD dissertation at the national university level.
PhD candidate Le Hong Lam expressed heartfelt gratitude to Assoc. Prof. Nguyen Ha Nam (supervisor), thanked the evaluation council, and expressed appreciation to the leadership of the Information Technology Institute, the Board of Directors of the University of Danang, and colleagues, friends, and especially his family for their constant encouragement and support, enabling him to successfully complete and defend his PhD dissertation.
Below are some images from the dissertation defense:
