编辑:时间:2022-02-28 16:06:12 浏览次数:
On February 19th, a statistical seminar jointly organized by the School of Statistics of JUFE and the School of Natural Sciences of Chonnam National University was successfully held. 3 teachers and 4 postgraduates from Chonnam University and 3 teachers and 4 doctoral students from the School of Statistics of JUFE exchanged reports on their recent work. The seminar was divided into two sessions. The morning session was chaired by Prof. Jaesik Jeong from Chonnam National University, and the afternoon session was presided over by Prof. Liu Xiaohui, Deputy Dean of the School of Statistics. Some teachers and postgraduates from the two universities participated in the joint seminar.
At the beginning of the meeting, Jansun Baek, a senior statistics professor at Chonnam University, gave a report titled Variational autoencoder with Manly transformation for clustering high-dimensional data. By analyzing the shortcomings of existing generative models, Professor Baek introduced a new class of high-dimensional clustering methods that can handle thick tails, are robust to outliers, and are conducive to display with low-dimensional visualization tools, and demonstrate their possibility of practical application.
Later, Professor Jeong-Soo Park gave a report titled Future projection of extreme rainfall in Korea: Determining shape parameters in a climate multi-model ensemble.The report describes a model weighting method that takes into account both model performance and independence. In the report, Professor Jeong-Soo Park introduced in detail how to reduce the computational time of the method by a simple method of selecting two shape parameters based on chi-square statistics and entropy.Finally, Professor Jeong-Soo Park also introduced the application of 21 CMIP6 (Coupled Model Intercomparison Project Phase 6) models for five climate variables in East Asia in obtaining future predictions of extreme rainfall on the Korean Peninsula.
Then, Professor Kyeong-Hwan Lee gave a report titled The new agriculture paradigm in Korea, digital agriculture, introducing the latest work done by his team in digital agriculture technology and its applications. In recent years, Prof. Kyeong-Hwan Lee's team has used technologies including sensors, ICT, big data, artificial intelligence (AI), unmanned aerial vehicles (UAV), and robotics in the Digital Agriculture System (DAS). DAS, introduced by Prof. Kyeong-Hwan Lee, enables farmers and other stakeholders in the agricultural value chain to increase production efficiency and facilitates both producers and consumers by sharing the information collected during agricultural production.
In the morning student report stage, statistics graduate students from Chonnam University, Yire Shin, Somi Cha, Seunghyeon Kim, and Jinwon Heo respectively gave reports entitled Modeling climate extremes using the r-largest four parameter kappa distribution and Extracting falling objects on the road from accident reports using a NLP model-based approach”, “A New Deep Learning Algorithm for Detecting the Lag Effect of Meteorological factors on Hospital Emergency Visits for Suicide Attempts” and “Variational autoencoder with gamma mixture model for clustering right-skewed data”.
In the JUFE report field in the afternoon, Mr. Ma Haiqiang first gave a report entitled Pseudo-Bayesian Classified Mixed Model Prediction, introducing a new classification mixed model prediction (CMMP) model, namely pseudo-Bayesian CMMP. In the report, Mr. Ma Haiqiang focused on the consistent optimality of the model in matching group indexes and predicting mixed effects related to new observations, and demonstrated the theoretical results and applications of related methods. The relevant results of Mr. Ma Haiqiang have been published online by JASA.
In the second report stage, Mr. Shan Qingsong gave a report entitled Have China's Poverty Alleviation Policies Improved Household Income? Evidence from Survey Data.This paper introduces the principle and idea of a class of copula entropy clustering trees that can extract poverty alleviation policy information from survey data, and introduces related applications. Shan Qingsong's report is mainly based on his latest collaborative paper published in Management World.
Shen Jingyi, Liu Yuzi, Peng Duyun and Liu Jun, doctoral students of statistics from JUFE, shared the reports entitledMeasurement of Chinese Export Green Trade Level and Analysis of Influencing Factors under the Dual Circulation Development Pattern and A Unified test for the Intercept of a Predictive. Regression Model, Does the Green Finance Pilot Reduce the Energy Consumption Intensity? Empirical Evidence from 30 Provinces in China, Statistical Decision-Making Theory and Method: Towards regional green development level evaluation.
In this joint academic conference, the reports of relevant teachers of the School of Statistics of JUFE fully demonstrated the depth and breadth of scientific research in the statistics department of JUFE. The solid report content and fluent English report of the statistics doctoral students of JUFE also fully demonstrated the good scientific research literacy and international exchange potential of the statistics doctoral students of JUFE.
Chonnam National University is a national university in Jeollanam-do, South Korea.It is a key construction university in South Korea's BK21 project.
Chonnam National University originated in 1909, its predecessor was Gwangju Agricultural University, and has made extensive achievements in digital agriculture. Chonnam National University in South Korea is a cooperative unit of the Rural Revitalization Project of the Foundation Remaining Committee of JUFE. The School of Natural Sciences and the School of Statistics of JUFE have a good cooperative relationship and plan to further deepen and further cooperate with each other in personnel training and scientific research.