Zhiyuan Liu
- +86 188 1153 9603
- zhiyuanliu0903@ruc.edu.cn
- Beijing, China
- Download My CV
Zhiyuan Liu (刘之源) is a final-year graduate student specializing in Management Science and Engineering at the School of Information, Renmin University of China. Her research interests revolve around Business Analytics, Social Media Analytics, Multimodal Data Analysis, and Artificial Intelligence. She employs techniques such as Data Mining, Natural Language Processing, Machine/Deep Learning, and Econometrics/Statistical Analysis to explore these areas.
Before her graduate study, Zhiyuan Liu majored in Computer Science and Technology and minored in Finance at Renmin University of China. Additionally, she developed a keen interest in Psychology during that period. This interdisciplinary learning experience has laid a solid foundation for her to conduct efficient research in multiple fields.
Publications
Journal Publications
Liu, Z., Xu, W., Zhang, W., & Jiang, Q. (2023). An emotion-based personalized music recommendation framework for emotion improvement. Information Processing & Management, 60(3), 103256.
Conference Presentations
Liu, Z., Liu, Y., Yan, Z., Yi, Z., & Zhang, W. (2023). Visualization performance evaluation of analyst reports. In The 16th China Summer Workshop on Information Management (CSWIM), Changsha.
Liu, Y., Liu, Z., & Zhang, W. (2023). Integrating the IPO roadshow performance for companies' business success prediction. In The 8th International Conference on Smart Finance (ICSF), Dubai.
Zhang, W., Liu, Z., Liu, Y., Yan, Z., & Yi, Z. (2023). The power of visualization on making high-quality business analytical reports: A multi-dimensional analysis. In INFORMS Annual Meeting 2023, Phoenix.
Working Papers
A fusion method to predict IPO success incorporating CEO's road show performance. Targeted Journal: INFORMS Journal on Computing. With Liu, Y. and Zhang W.
How Physicians Gain Trust through Photos? A Multi-Dimensional Study on Impression Management in Online Healthcare Consultation. With Wang, X., Huang, T., Zhang, W.
Research Experience
Idea Plagiarism in Short Videos: A Semantic-based Multi-model Method
Student Researcher, Advised by Assoc. Prof. Wenping Zhang, Renmin Information School
Develop a two-stage framework for idea plagiarism detection in short videos
Extract semantic features as unique identifiers for identifying suspicious video
Utilize multi-model techniques to compare semantic similarities between text, image and video data for idea plagiarism detection
Multi-dimensional Evaluation for Analytical Reports
Team Leader, Advised by Prof. Zhihong Yi, Renmin Business School, and Assoc. Prof. Wenping Zhang, Renmin Information School
Construct an innovative model to evaluate the information quality of analytical reports
Develop algorithms of text analysis and image understanding to calculate some of the indicators; establish empirical study on real data from Wind to verify the effectiveness of our model
Lead a 5-person research team of the Information School in collaboration with students from the Business School
Multi-dimensional Study on Information Management in Online Healthcare Consultation
Research Assistant, Advised by Assoc. Prof. Wenping Zhang, Renmin Information School, and Prof. Huang Tao, Peking University
Conduct a systematic investigation into the impacts of physicians' photographs on patients' whole decision-making processes, namely search, selection, and evaluation online
Construct a four-dimensional facial impression model and verify the effectiveness of our proposed model on the data from one of the largest online healthcare platforms
Music Recommendation for Emotion Improvement
Student Researcher, Advised by Assoc. Prof. Wenping Zhang, Renmin Information School
Designed an LSTM-based model to select the most suitable and helpful music based on users’ moods in the previous period and current emotion stimuli
Applied empirical experiments and user studies to prove that our novel framework is precise and helpful
Short Video Understanding Based on Data Augmentation
Student Researcher, Advised by Prof. Wei Xu and Assoc. Prof. Wenping Zhang, Renmin Information School
Proposed a new video data augmentation method to better identify the features of short videos
Brought a certain improvement of 2% in the accuracy of short video understanding
Work Experience
Machine Learning Engineer
Applied Mask R-CNN model to food category identification
Constructed and maintained a food database that promoted model updates and accuracy improvements