Using Machine Learning to evaluate forensic evidence
China is developing a system that uses artificial intelligence and big data to detect financial fraud, including tricks unknown to most tax inspectors.
Researchers believe that the traditional tax system uses all available data too effectively. Therefore, for three years now, Aisino has been working with scientists from Harbin Institute of Technology and Beijing University of Posts and Telecommunications to develop an automated solution..
According to the team, during this time about 300 thousand rubles. state inspectors took part in training the algorithms of the system. This program is capable of detecting more than 95% of tax crimes, including those that people would not even be able to notice..
The developed mechanism is integrated directly into the software used by the State Tax Service of China. The administration has not yet disclosed the details of the project, but the effectiveness of the system is known to be confirmed by the effectiveness of pilot programs in economic centers in the east of the country. Although the government has not yet approved a full-scale launch..
The new system can be connected to all government databases, including information on property, goods, international trade and business registration. This allows technology to automatically deploy, whether companies and individuals have misrepresented information in declarations, as well as find new methods of evasion, previously unknown to the authorities.
In the process of work, the system learns itself, updating new circuits..
Researchers expect that artificial intelligence will reduce the influence of the human factor in the field of taxation, and an increase in treasury revenues will determine the country’s economic growth..
Despite the rules of new technologies, there are new ways to bypass them.. Representatives of the usual print program with the face of another person have recently appeared…
text: Ilya Bauer, photo: Reuters
Artificial Intelligence (AI) and its applications in Forensics [e-forensics 2020]