Forensic accountants today are required to possess a number of characteristics and skills that are derived from a range of disciplines. They must be inquisitive, able to delve into motivations, operations and justific...Forensic accountants today are required to possess a number of characteristics and skills that are derived from a range of disciplines. They must be inquisitive, able to delve into motivations, operations and justifications underlying the behavior of white collar criminals. They must be able to simplify technical accounting information and convey it effectively orally and in writing to a court or tribunal; be able to trace information via information technology tools; have an understanding of the adversarial legal system and the rules of evidence. Universities offering forensic accounting courses in various formats are on the increase. The aim of this paper is to investigate the subject areas that any course specializing in forensic accounting needs to cover.展开更多
The feature analysis of fraudulent websites is of great significance to the combat,prevention and control of telecom fraud crimes.Aiming to address the shortcomings of existing analytical approaches,i.e.single dimensi...The feature analysis of fraudulent websites is of great significance to the combat,prevention and control of telecom fraud crimes.Aiming to address the shortcomings of existing analytical approaches,i.e.single dimension and venerability to anti-reconnaissance,this paper adopts the Stacking,the ensemble learning algorithm,combines multiple modalities such as text,image and URL,and proposes a multimodal fraudulent website identification method by ensembling heterogeneous models.Crossvalidation is first used in the training of multiple largely different base classifiers that are strong in learning,such as BERT model,residual neural network(ResNet)and logistic regression model.Classification of the text,image and URL features are then performed respectively.The results of the base classifiers are taken as the input of the meta-classifier,and the output of which is eventually used as the final identification.The study indicates that the fusion method is more effective in identifying fraudulent websites than the single-modal method,and the recall is increased by at least 1%.In addition,the deployment of the algorithm to the real Internet environment shows the improvement of the identification accuracy by at least 1.9%compared with other fusion methods.展开更多
文摘Forensic accountants today are required to possess a number of characteristics and skills that are derived from a range of disciplines. They must be inquisitive, able to delve into motivations, operations and justifications underlying the behavior of white collar criminals. They must be able to simplify technical accounting information and convey it effectively orally and in writing to a court or tribunal; be able to trace information via information technology tools; have an understanding of the adversarial legal system and the rules of evidence. Universities offering forensic accounting courses in various formats are on the increase. The aim of this paper is to investigate the subject areas that any course specializing in forensic accounting needs to cover.
基金supported by Zhejiang Provincial Natural Science Foundation of China(Grant No.LGF20G030001)Ministry of Public Security Science and Technology Plan Project(2022LL16)Key scientific research projects of agricultural and social development in Hangzhou in 2020(202004A06).
文摘The feature analysis of fraudulent websites is of great significance to the combat,prevention and control of telecom fraud crimes.Aiming to address the shortcomings of existing analytical approaches,i.e.single dimension and venerability to anti-reconnaissance,this paper adopts the Stacking,the ensemble learning algorithm,combines multiple modalities such as text,image and URL,and proposes a multimodal fraudulent website identification method by ensembling heterogeneous models.Crossvalidation is first used in the training of multiple largely different base classifiers that are strong in learning,such as BERT model,residual neural network(ResNet)and logistic regression model.Classification of the text,image and URL features are then performed respectively.The results of the base classifiers are taken as the input of the meta-classifier,and the output of which is eventually used as the final identification.The study indicates that the fusion method is more effective in identifying fraudulent websites than the single-modal method,and the recall is increased by at least 1%.In addition,the deployment of the algorithm to the real Internet environment shows the improvement of the identification accuracy by at least 1.9%compared with other fusion methods.
