Fei Gao

Dr. Fei Gao

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 May 2022 in Security and Communication Networks
Verified
Journal Article
Open access
PDF file
KCKai Cong
Kai Cong
TLTao Li
Tao Li
Beibei LiBeibei Li
Beibei Li
ZGZhan Gao
Zhan Gao
YXYanbin Xu
Yanbin Xu
FGFei Gao
Fei Gao
The Bidirectional Encoder Representations from Transformers (BERT) technique has been widely used in detecting Chinese sensitive information. However, existing BERT-based frameworks usually fail to emphasize key entities in the texts that contribute significantly to knowledge inference. To meet this gap, we propose a BERT and knowledge graph-based novel framework to detect Chinese sensitive information (named KGDetector). Specifically, we first train a pretrained knowledge graph-based Chinese entity embedding model to characterize entities in the Chinese textual inputs. Finally, we propose an effective framework KGDetector to detect Chinese sensitive information, which employs the knowledge graph-based embedding model and the CNN classification model. Extensive experiments on our crafted Chinese sensitive information dataset demonstrate that KGDetector can effectively detect Chinese sensitive information, outperforming existing baseline frameworks.

2 April 2024 in Forests
Verified
Journal Article
Open access
PDF file
FGFei Gao
Fei Gao
Xiaohui SuXiaohui Su
Xiaohui Su
YCYuling Chen
Yuling Chen
BWBaoguo Wu
Baoguo Wu
YTyingze tian
yingze tian
TLTao Li
Tao Li
The National Forestland “One Map” applies the boundaries and attributes of sub-elements to mountain plots by means of spatial data to achieve digital management of forest resources. The change detection and analysis of forest space and property is the key to determining the change characteristics, evolution trend and management effectiveness of forest land. The existing spatial overlay method, rasterization method, object matching method, etc., cannot meet the requirements of high efficiency and high precision at the same time. In this paper, we investigate a fast algorithm for the detection of changes in “One Map”, taking Sichuan Province as an example. The key spatial characteristic extraction method is used to uniquely determine the sub-compartments. We construct an unbalanced quadtree based on the number of maximum leaf node elements (NLNE Quad-Tree) to narrow down the query range of the target sub-compartments and quickly locate the sub-compartments. Based on NLNE Quad-Tree, we establish a change detection model for “One Map” (NQT-FCDM). The results show that the spatial feature combination of barycentric coordinates and area can ensure the spatial uniqueness of 44.45 million sub-compartments in Sichuan Province with 1 m~0.000001 m precision. The NQT-FCDM constructed with 1000–6000 as the maximum number of leaf nodes has the best retrieval efficiency in the range of 100,000–500,000 sub-compartments. The NQT-FCDM shortens the time by about 75% compared with the traditional spatial union analysis method, shortens the time by about 50% compared with the normal quadtree and effectively solves the problem of generating a large amount of intermediate data in the spatial union analysis method. The NQT-FCDM proposed in this paper improves the efficiency of change detection in “One Map” and can be generalized to other industries applying geographic information systems to carry out change detection, providing a basis for the detection of changes in vector spatial data.

20 March 2020
Verified
Book Chapter
Closed access
QSQiancheng Shan
Qiancheng Shan
ELEnshen Long
Enshen Long
Yin ZhangYin Zhang
Yin Zhang
FGFei Gao
Fei Gao
ZWziyun wang
ziyun wang