Mingzhao Li

Understanding the skills and proficiency levels required for various roles is crucial for effective workforce planning, learning and development. In this paper, we propose a robust skill proficiency modeling framework that offers a structured method to help describe, assess and develop proficiency in key skills, facilitating individuals' career pathways and aiding organizations in talent management and adaptability. We first design a skill proficiency description pipeline, which generates statements describing the requirements at each proficiency level of a skill. Following this, we build a skill proficiency by occupation model using large-scale job ad data to help organizations and individuals understand the skill proficiency requirements for different roles. Finally,we design a visual analytics system, based on a real-world career pathway scenario, to demonstrate the practical usefulness and effectiveness of our framework. A demo video is available at www.dropbox.com/scl/fi/nd0f3vi03n12g4y0sluaw/cikm24_demo.mp4?rlkey=55vya144q5ftai1uqqaubr5u5.

In this paper, we propose a novel facility relocation problem where facilities (and their services) are portable, which is a combinatorial search problem with many practical applications. Given a set of users, a set of existing facilities, and a set of potential sites, we decide which of the existing facilities to relocate to potential sites, such that two factors are satisfied: (1) facility exposure: facilities after relocation have balanced exposure, namely serving equivalent numbers of users; (2) user convenience: it is convenient for users to access the nearest facility, which provides services with shorter travel distance. This problem is motivated by applications such as dynamically redistributing vaccine resources to align supply with demand for different vaccination centers, and relocating the bike sharing sites daily to improve the transportation efficiency. We first prove that this problem is NP-hard, and then we propose two algorithms: a non-learning best response algorithm () and a reinforcement learning algorithm (). In particular, the best response algorithm finds a Nash equilibrium to balance the facility-related and the user-related goals. To avoid being confined to only one Nash equilibrium, as found in the method, we also propose the reinforcement learning algorithm for long-term benefits, where each facility is an agent and we determine whether a facility needs to be relocated or not. To verify the effectiveness of our methods, we adopt multiple metrics to evaluate not only our objective, but also several other facility exposure equity and user convenience metrics to understand the benefits after facility relocation. Finally, comprehensive experiments using real-world datasets provide insights into the effectiveness of the two algorithms in practice.

The study of tourism destination images is of great significance in the tourism discipline. Tourism user-generated content (UGC), i.e., the feedback on tou

We aimed to evaluate whether or not time lapse selection was beneficial for the cleavage-stage embryo transfers. The study included 838 infertile women with good ovarian reserve (obtaining more than 8 oocytes) from January 2018 to August 2019. Based on the transferred embryos with different grades (grade I, II and III), the patients were divided into day 3 selection with conventional morphology (CM) and day 3 selection with time lapse (TL) groups. For the grade I and II embryos, we observed that CM and TL had similar implantation, clinical pregnancy and ongoing pregnancy (p > .05) rates. For the grade III embryos, we observed that CM group showed slightly lower implantation (36.74 versus 41.03%, p = .261) and clinical pregnancy (56.82 versus 64.10%, p = .182) rates than TL group. CM group showed significantly lower ongoing pregnancy (47.35 versus 59.83%, p = .025) rate than TL group. And we observed that CM group had significantly higher blastulation (38.93 versus 26.61%, p = .019) rate than TL group. We concluded that TL selection was beneficial to the patients with no good-quality embryos in the first cleavage-stage embryo transfers.