Prof. Defu Zhang
Research Area: Big Data & AI
Speech Title: Machine learning for Stock forecasting
Abstract: This talk mainly introduces different models for stock forecasting, one model combines features selected by multiple feature selection techniques to generate an optimal feature subset and then use a deep generative model to predict future stock movements. Another model is for a turning point prediction method of stock price based on RVFL-GMDH and chaotic time series analysis. The turning indicator of time series is computed firstly; then, by applying the RVFL-GMDH model for the turning point prediction of the stock price. Computation results are reported.
Assoc. Prof. Vasilii Erokhin
Harbin Engineering University
Research Area: International trade, globalization, trade integration and liberalization, sustainable development, food security issues with a focus on emerging markets, developing countries, and economies in transition
Speech Title: Economic Prospects of Artificial Intelligence in the Big Data Environment
Abstract: According to many experts, technical and technological progress will soon allow businesses to replace most jobs with computers, machines, and other digital technologies. Artificial intelligence is one of the most rapidly developing solutions today. Big data technologies greatly support the spread and improvement of artificial intelligence across various sectors. However, it is difficult to predict how artificial intelligence will actually develop and what possible consequences it may bring for the future of the labor market and the well-being of people. In my speech, I am going to discuss the potential benefits of artificial intelligence for economic development, as well as potential threats of intelligence-related technologies to various industries and markets.
Assoc. Prof. Min Hou
Zhejiang Gongshang University
Research Area: Advertising; Bilateral markets;, Internet finance, Consumer behavior
Speech Title: The Impact of Pragmatic Markers on Perceived Usefulness of Online Reviews: An Analysis Based in the Chinese E-commerce Context
Abstract: This study uses text analysis to explore the impact of pragmatic markers (classified as straightforward, elaborative, motive, assertive, evaluative, caution, and additional markers) on the perceived usefulness of reviews. To see whether and how Pragmatic Markers affect the perceived usefulness of reviews. In addition, online reviews also have the special context of mixed reviews. We captured online reviews from a shopping website in China, JD.com. According to the characteristics of pragmatic markers, we constructed a logistic regression model to test the research hypotheses. We found that pragmatic markers had a significant positive impact on the perceived usefulness of online reviews. In addition, straightforward, elaborative, motive, and assertive markers had significant and positive effects on the perceived usefulness of online reviews. The study not only elucidates the factors influencing online reviews’ usefulness from a different perspective, it also provides a reference for e-commerce platforms and reviewers.
Assoc. Prof. ZiqiangZeng
Research Area：Data-driven decision making; multistage decision making and engineering optimization
Speech Title: Digital Twin Platform for Traffic Emission Analysis of Chengdu City
Abstract: In recent years, China's regional atmospheric environmental problems have become increasingly prominent. It has become an important national strategy to control urban traffic pollution emissions and improve air quality. The construction of traffic emission big data analysis platform can realize low-cost real-time analysis and visualization of traffic emission data. It can also evaluate the environment and energy-saving benefits of traffic planning and decision-making, guide popular low-carbon travel, effectively promote the specific implementation of intelligent traffic management policy, eliminate the data barrier of industrial docking, promote the development of green transportation industry chain, revitalize the upstream and downstream resources of related industries, and bring new vitality to a city’s economy. Based on the data resources related to traffic emissions in Chengdu, this study constructed the basic framework and algorithm of the digital twin platform for traffic emission analysis, and proposed the traceability analysis of traffic pollution emissions on road sections, that is, based on the road section emission prediction model, through big data analysis to locate possible high pollution road sections or regions, and dynamically analyze the formation, transmission and diffusion of traffic pollution emissions. On the basis of judging the formation of high pollution, this study further analyzes the causes and trends of high pollution, and provides a reliable basis for management decisions such as environmental accountability, punishment and rectification. Finally, from the perspectives of government, enterprises and public, this study discusses the construction of ecological chain system of related data application.