Systems and methods for detecting data anomalies by analysing morphologies of known and/or unknown cybersecurity threats
AssigneeFlexxon Pte. Ltd.
InventorsNizar Bouguerra, Chan Mei Ling
Life of this patent
Estimating the term…
Abstract
Systems and methods for detecting anomalous data files and preventing detected anomalous data files from being stored in a data storage. In particular, the systems and methods detect anomalous data files by dividing each data file into blocks of data whereby entropy values are obtained for each block of data and this information is collated and subsequently used in a machine learning model to ascertain the security level of the data file.
Claims22
A system for detecting data anomalies in a received data object, the system comprising:
- a processing unit;
- and
- a non-transitory media readable by the processing unit, the media storing instructions that when executed by the processing unit, causes the processing unit to:
- determine a security posture of the data object based on a digital signature and a file type of the data object;
- generate a type-security-platform (TSP) lookup table based on the security posture and features of the data object associated with the security posture, and generate an obfuscation value and a forensic value for the received data object based on the TSP lookup table;
- generate disassembled values or interpreted values for the data object;
- compute a result value for each block of the received data object, whereby the result values for each of the respective blocks of the received data object are generated based on the disassembled or interpreted values, an obfuscation value and a forensic value associated with the respective block of the received data object;
- create a data model based on all the result values of the data object;
- and
- process the data model using an artificial intelligence (AI) algorithm to determine if the data object contains data anomalies.
A method for detecting data anomalies in a received data object using an Artificial Intelligence (AI) module, the method comprising:
- determining, using an analyser module provided within the AI module, a security posture of the data object based on a digital signature and a file type of the data object;
- generating, using the analyser module and a detector module provided within the AI module, a type-security-platform (TSP) lookup table based on the security posture and features of the data object associated with the security posture, and generating an obfuscation value and a forensic value for the received data object based on the TSP lookup table;
- generating, using a disassembling and interpreting module provided within the AI module, disassembled values or interpreted values for the data object;
- computing, using a block structuring module provided within the AI module, a result value for each block of the received data object, whereby the result value for each of the respective blocks of the received data objects is generated based on the disassembled or interpreted values, an obfuscation value and a forensic value associated with the block of the received data object;
- creating, using a model generator module provided within the AI module, a data model based on all the result values of the data object;
- and
- processing, using an AI threat module provided within the AI module, the data model using an artificial intelligence (AI) algorithm to determine if the data object contains data anomalies.
Dependencies are inferred from the claim text.