Research
Research
Towards a Digital Age: Exploring the Digital Bonds
Multi-dimensional Housing Inequality Index: The Provincial Evidence from China
Junhua Chen*, Shufan Ma, and Na Liu.
Social Indicators Research, 2022
This research constructed a multi-dimensional inequality index to measure China's housing inequality from three dimensions: housing living conditions, housing wealth, and housing welfare based on household registration (hukou). Using cross-sectional data from the four-times China Household Finance Survey, the article showed the degree of China's housing inequality has an overall upward trend from 2011 to 2017. Furthermore, this research calculated the provincial housing inequality and conducted the regression analysis to investigate the factors that affect housing inequality in each province and suggests the central government focus on housing welfare and reform the hukou system in the future to alleviate housing inequality and the contradiction of housing interests
Lu Wei*, Shufan Ma, Maoze Wang
Electronic Commerce Research, 2023
Online reviews are essential to consumers' decision-making when purchasing products on e-commerce platforms. Most of the existing research conducts sentiment analysis on online reviews but needs to pay more attention to the information features of the text of online reviews. Based on Chinese review texts, this study collected 18,131 online clothing review information and applied Latent Dirichlet allocation (LDA) to divide the review texts into nine topics. Then, the informative characteristics of online reviews were evaluated using review length, readability, redundancy, and specificity indicators. We also investigate the relationship between review text informative features and review sentiment and verify the robustness of the results using different regression models. Our research will help e-commerce platforms construct general review guidelines to improve consumer satisfaction.
International Carbon Financial Market Prediction Using PSO and SVM Algorithm
Junhua Chen, Shufan Ma, and Ying Wu*
Journal of Ambient Intelligence and Humanized Computing, 2021
This research presented a novel approach, PSO-SVM, by combining support vector machine (SVM) and particle swarm optimization (PSO) algorithm and provided the optimal parameters for SVM to improve the prediction performance of European Allowance (EUA) carbon emission futures. This research used the realistic trading dataset containing 30,762 EUA futures closing prices to effectively predict extreme price fluctuations and overcome the problem of high prediction error caused by parameter constraints.
Note: * means the corresponding author.
Mass media, medical insurance and heterogeneous growth of companies: Evidence from the “Under the Dome” in China
Shufan Ma, Weizeng Sun
Land Financing Mode and Urban Total Factor Productivity Growth: Evidence from Chinese Cities
Junhua Chen, Shufan Ma, Yi Yang