Zhihong Zhao

Prof. Zhihong Zhao

Beijing Institute of Technology, Zhuhai, Zhuhai, Guangdong, China
orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
19 August 2022 in Entropy
Verified
Journal Article
Open access
PDF file
ZZZhihong Zhao
Zhihong Zhao
TBTongyuan Bai
Tongyuan Bai
This paper proposes a new method that can identify and predict financial fraud among listed companies based on machine learning. We collected 18,060 transactions and 363 indicators of finance, including 362 financial variables and a class variable. Then, we eliminated 9 indicators which were not related to financial fraud and processed the missing values. After that, we extracted 13 indicators from 353 indicators which have a big impact on financial fraud based on multiple feature selection models and the frequency of occurrence of features in all algorithms. Then, we established five single classification models and three ensemble models for the prediction of financial fraud records of listed companies, including LR, RF, XGBOOST, SVM, and DT and ensemble models with a voting classifier. Finally, we chose the optimal single model from five machine learning algorithms and the best ensemble model among all hybrid models. In choosing the model parameter, optimal parameters were selected by using the grid search method and comparing several evaluation metrics of models. The results determined the accuracy of the optimal single model to be in a range from 97% to 99%, and that of the ensemble models as higher than 99%. This shows that the optimal ensemble model performs well and can efficiently predict and detect fraudulent activity of companies. Thus, a hybrid model which combines a logistic regression model with an XGBOOST model is the best among all models. In the future, it will not only be able to predict fraudulent behavior in company management but also reduce the burden of doing so.
1

19 May 2024 in Sustainability
Verified
Journal Article
Open access
PDF file
JSJianwei Shi
Jianwei Shi
TBTongyuan Bai
Tongyuan Bai
ZZZhihong Zhao
Zhihong Zhao
Huachun TanHuachun Tan
Huachun Tan
This research investigates the crucial role of transportation infrastructure in influencing economic activity, thus employing advanced econometric methods including Moran’s I index, LM, Hausman, and LR tests to ensure analytical accuracy and select the appropriate spatial model. Our findings reveal that freight volumes across road, waterway, and civil aviation significantly enhance economic activity by bolstering domestic trade, industrial production, and supply chains. Conversely, the impact of passenger turnover is comparatively minor, although it still contributes to labor mobility and urban accessibility. This study highlights the need for strategic investment in transportation infrastructure and efficient public transport systems to foster economic growth and sustainable development. We recommend that policymakers focus on optimizing transportation networks and integrating intelligent transport technologies to boost economic competitiveness and societal well-being. This analysis not only sheds light on the direct economic impacts of transportation but also underscores the broader social implications, thus advocating for a holistic approach to transportation planning and policymaking.

25 November 2022
Verified
Conference Paper
Open access
ZZZhihong Zhao
Zhihong Zhao
JYJieshun You
Jieshun You
YCYunong Cui
Yunong Cui
The use of computer vision techniques to distinguish fresh from stale fruit has gained widespread acceptance in recent years. However, the scarcity of datasets in this field has made it more difficult for many of the research results to be accessible in real-life situations. The core idea of this paper is to improve the generalization of the training set and the performance of the classifier by augmenting the fruit and vegetable image dataset with neural networks. This paper will investigate the applicability of the established Conditional Generative Adversarial Nets (CGAN) in this domain and discover problems such as mode collapse. The contribution of this paper is to combine existing structures and propose novel adversarial generative networks for the fruit and vegetable image, including Conditional Wasserstein GAN with Gradient Penalty (CWGAN-GP) and Auxiliary Classifier Wasserstein GAN with Gradient Penalty (ACWGAN-GP). In addition, a pre-trained models with Spinal fully-connected layer and ProgressiveSpinal fully-connected layer was used to freshness classify. It was found that ACWGAN-GP, which combines multiple network structures, not only produced more realistic and diverse images in the fruit and vegetable domain, but also maintained the learning speed of a single model during the training. In addition, by adding synthetic images to the training, the fruit and vegetable freshness classifier achieves 100% accuracy on a specified validation set.