Analyzing Career Trajectories of Qing Dynasty Civil Servants with Data Mining and Machine Learning

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering

Final Year Thesis Oral Defense

Title: "Analyzing Career Trajectories of Qing Dynasty Civil Servants with 
        Data Mining and Machine Learning"

by

WANG Yanbang

Abstract:

Nowadays, many large companies and organizations keep track of their 
personnel appointment records. One problem of interest is what we call the 
"stepping-stone position" problem: within the hierarchical organization of 
positions, can we identify certain positions whose servants possess the 
best potential of promotion in the future? To tackle the problem, we first 
proposed two baseline solutions, followed by a more refined graph-based 
model and its paired algorithm, PosRank. We experimented PosRank in 
Jinshen Lu Database, and worked out a global ranking of position's 
potentials. The results were validated by domain experts, who gave very 
positive feedback. In-depth follow-up analysis, including BFS-based 
subgraph analysis, geographical analysis, and ranking reconstruction using 
partial data, were also conducted to further exploit relevant knowledge as 
well as to demonstrate the robustness of our method. To our best 
knowledge, we are among the first to formulate the problem of ranking the 
potentials of positions, and to actually generate the numerical values. 
The study also serves as an illuminating demonstration that big data 
analytics can serve as a complementary element to traditional historical 
research.


Date            : 3 May 2019 (Friday)

Time            : 15:10 - 15:50

Venue           : Room 4621 (near lifts 31/32), HKUST

Advisor         : Prof. GOLIN Mordecai J.

2nd Reader      : Prof. KELLER Barbara Franziska