Yuan Cheng

Yuan Cheng

Auto-generated profile, based on publicly available information

Yuan Cheng

Career Science Lab, BOSS Zhipin, Beijing, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

7 May 2024 in IEEE Transactions on Knowledge and Data Engineering
Verified
Journal Article
Closed access
ZZZhi Zheng
Zhi Zheng
XHXiao Hu
Xiao Hu
ZQZhaopeng Qiu
Zhaopeng Qiu
YCYuan Cheng
Yuan Cheng
SGShanshan Gao
Shanshan Gao
HXHui Xiong
Hui Xiong
Recent years have witnessed the rapid development of online recruitment platforms, which provide a convenient way for matching job seekers and recruiters. Indeed, this is a reciprocal recommendation problem which needs to consider the preferences of both job seekers and recruiters simultaneously. Existing studies mainly focus on building recommendation models based on text matching or collaborative filtering methods. However, we propose that these methods are limited and insufficient, since the abundant multi-typed bilateral behaviors among users have been largely ignored. Therefore, in this paper, we propose a novel BilAteral Multi-BehaviOr mOdeling (BAMBOO) method for reciprocal recommendation in online recruitment, which can model the multi-typed interactions from expectation perspective and competitiveness perspective . Specifically, for the expectation perspective, we propose to format the historical behaviors of different users as bilateral multi-behavior sequences, and we utilize a transformer-based model to learn the representations of what the users want to obtain. For the competitiveness perspective, we propose to construct a bilateral interaction heterogeneous graph to describe the entire recruitment market, and further utilize a heterogeneous graph transformer-based model to learn the representations of what the users can obtain. Moreover, we utilize contrastive learning methods to enhance these two modules. Furthermore, we propose to decompose the matching probability into the product of two parts, and we train our model based on a multi-task learning strategy. We conduct both offline experiments on real-world datasets and online A/B test, and the experiment results validate the effectiveness of our BAMBOO model compared with several state-of-the-art baseline methods.

18 March 2025 in Humanities and Social Sciences Communications
Verified
Journal Article
Open access
PDF file
YCYihang Cheng
Yihang Cheng
ZYZhaoqi Yang
Zhaoqi Yang
YCYuan Cheng
Yuan Cheng
YGYong Ge
Yong Ge
HZHengshu Zhu
Hengshu Zhu
Recently, the strong artificial intelligence-based augmented capability enables the autonomous completion of traditional jobs devoid of human intervention, impacting labor markets. However, the underlying mechanisms have not been explored enough in prior research. In this research, we propose a computational model, focusing on the interplay between world knowledge networks and organizational knowledge sets, along with external labor market conditions. This model incorporates dynamic knowledge creation behaviors and is validated using a substantial dataset from a leading online recruitment platform in China, featuring over 20 million job postings and 1 million skill-related keywords. The results demonstrate that swift knowledge search and emulating knowledge within existing jobs are the main methods in the early developmental stage of organizations, accounting for about 75% of all simulation samples and forming the initial job evolution. As organizations progress, although fine-tuning the knowledge within existing jobs still remains significant, the intensity of knowledge search declines significantly, and the intensity of knowledge reuse surpasses that of knowledge search, reaching ~1.5 times its intensity during the stable phase. We also perform several parameter experiments and a case study to illustrate how jobs evolve in the labor market with different characteristics. The robustness tests demonstrate the model’s resilience across different simulation environments and organization strategies. Our study underlies the mechanisms of job evolution from the organizational level and provides empirical evidence and insights into the job evolution dynamics within knowledge networks.

8 November 2023 in Pattern Recognition Letters
Verified
Journal Article
Closed access
YYYaqi Yang
Yaqi Yang
YZYang Zhao
Yang Zhao
YCYuan Cheng
Yuan Cheng