Around 70,000 to 80,000 electronic documents were received through the system every day and about 2,000-3,000 of them were suspected to have inaccurate information included or selected for manual check. For those suspected with inaccurate information document, officers concerned need to follow up by contacting the submitter through telephone or query through the system for clarification and correction. For those documents suspected to involve deceptive information, the case would be referred to law enforcement agencies for investigation. The project aims to enhance the existing system using artificial intelligence, machine learning and other technologies to identify and process the incoming electronic documents with suspected inaccurate information.
