Guangshun Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Guangshun Li has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
Guangsheng Ma
Harbin Engineering University
1 shared publication- Huiyu Zhou
University of Leicester, Leicester, United Kingdom
1 shared publication - Junhua Wu
School of Computer Science, Qufu Normal University, Rizhao, China
1 shared publication - Junliang Shang
School of Computer Science, Qufu Normal University, Rizhao, China
1 shared publication - Kui Lu
School of Computer Science, Qufu Normal University, Rizhao, China
1 shared publication - Peilin Li1 shared publication
Latest Publications
- Open Access
PFed-Signal: An ADR Prediction Model based on Federated Learning
The adverse drug reactions (ADRs) predicted based on the biased records in FAERS (U.S. Food and Drug Administration Adverse Event Reporting System) may mislead diagnosis online. Generally, such problems are solved by optimizing reporting odds ratio (ROR) or proportional reporting ratio (PRR). However, these methods that rely on statistical methods cannot eliminate the biased data, leading to inaccurate signal prediction. In this paper, we propose PFed-signal, a federated learning-based signal prediction model of ADR, which utilizes the Euclidean distance to eliminate the biased data from FAERS, thereby improving the accuracy of ADR prediction. Specifically, we first propose Pfed-Split, a method to split the original dataset into a split dataset based on ADR. Then we propose ADR-signal, an ADR prediction model, including a biased data identification method based on federated learning and an ADR prediction model based on Transformer. The former identifies the biased data according to the Euclidean distance and generates a clean dataset by deleting the biased data. The latter is an ADR prediction model based on Transformer trained on the clean data set. The results show that the ROR and PRR on the clean dataset are better than those of the traditional methods. Furthermore, the accuracy rate, F1 score, recall rate and AUC of PFed-Signal are 0.887, 0.890, 0.913 and 0.957 respectively, which are higher than the baselines.
Open AccessPFed-Signal: An ADR Prediction Model based on Federated Learning
The adverse drug reactions (ADRs) predicted based on the biased records in FAERS (U.S. Food and Drug Administration Adverse Event Reporting System) may mislead diagnosis online. Generally, such problems are solved by optimizing reporting odds ratio (ROR) or proportional reporting ratio (PRR). However, these methods that rely on statistical methods cannot eliminate the biased data, leading to inaccurate signal prediction. In this paper, we propose PFed-signal, a federated learning-based signal prediction model of ADR, which utilizes the Euclidean distance to eliminate the biased data from FAERS, thereby improving the accuracy of ADR prediction. Specifically, we first propose Pfed-Split, a method to split the original dataset into a split dataset based on ADR. Then we propose ADR-signal, an ADR prediction model, including a biased data identification method based on federated learning and an ADR prediction model based on Transformer. The former identifies the biased data according to the Euclidean distance and generates a clean dataset by deleting the biased data. The latter is an ADR prediction model based on Transformer trained on the clean data set. The results show that the ROR and PRR on the clean dataset are better than those of the traditional methods. Furthermore, the accuracy rate, F1 score, recall rate and AUC of PFed-Signal are 0.887, 0.890, 0.913 and 0.957 respectively, which are higher than the baselines.
- 1 May 2007
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1701-1705
A New Model K*TDG and High Level Exploring Algorithm
A new model K*TDG is proposed in this paper to solve the problem of complex data flow decomposing and exploring, where k indicates the compactness of the system parameters. In the following, the attributes of K*TDG and its primary operations are discussed, and a decomposing and exploring algorithm is proposed. In order to decrease the algorithm complexity, a group strategy on the degree of polynomials of complex components is shown. Experimental results indicate that using the algorithm and strategy in this paper, keeping the area and delay approximate optimization, the exploring space is decreased greatly.
- 1 May 2007
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1701-1705
A New Model K*TDG and High Level Exploring Algorithm
A new model K*TDG is proposed in this paper to solve the problem of complex data flow decomposing and exploring, where k indicates the compactness of the system parameters. In the following, the attributes of K*TDG and its primary operations are discussed, and a decomposing and exploring algorithm is proposed. In order to decrease the algorithm complexity, a group strategy on the degree of polynomials of complex components is shown. Experimental results indicate that using the algorithm and strategy in this paper, keeping the area and delay approximate optimization, the exploring space is decreased greatly.