Abstract: The invention is a method for securing a system including a plurality of entities and a repository storing usage patterns. The method comprises the following steps: for each of the entities involved in a transaction requested by a user, capturing a behavior data reflecting the dynamic behavior of said entity, generating a set of locality-sensitive-hashing hashes from each of said behavior data, computing a similarity score for each of the entities involved in the transaction, based on both said usage patterns and the set of generated locality-sensitive-hashing hashes, and based on a subset of the computed similarity scores, applying a preset security action to the transaction.

Abstract: A system configured to perform decision tasks carried out by a machine learning engine operates with a machine learning model, and includes a training component for improving the machine learning model, a device for carrying out decisions based on a set of input data, and an interaction interface for switching the machine learning model between training component and a device that includes a model attestation checker. The device performs acquiring input data, and ascertaining at least one machine learning model over the interaction interface. The model attestation checker performs checking if said machine learning model is trusted by a model attestation, and considering, for decision making, only those machine learning models that are trusted. The machine learning engine performs carrying out the decision task for input data by using a trusted machine learning model, and providing a result attestation for the decision output.

Abstract: The invention is a method for securing a system including a plurality of entities and a repository storing usage patterns. The method comprises the following steps: for each of the entities involved in a transaction requested by a user, capturing a behavior data reflecting the dynamic behavior of said entity, generating a set of locality-sensitive-hashing hashes from each of said behavior data, computing a similarity score for each of the entities involved in the transaction, based on both said usage patterns and the set of generated locality-sensitive-hashing hashes, and based on a subset of the computed similarity scores, applying a preset security action to the transaction.

Abstract: The invention relates to a method for sending data to at least one device. According to the invention, a data sending control server sends to at least one data storage server at least one predetermined rule or a first request for sending data to at least one data processing server. The data storage server sends, based upon the at least one predetermined rule or the first request for sending data, data to the data processing server. The data sending control server sends to the data processing server a second request for sending to the device the data received or to be received by the data processing server. The data processing server sends, based upon the second request for sending the data, the received data to the at least one device.

Abstract: The present invention relates to a method for authentication of a user using a user equipment, comprising an authentication engine for authenticating at least one user, said authentication engine being configured to operate with a local authentication model, wherein the method comprises the step for the user equipment of enhancing its local authentication model by at least one authentication factor, wherein said at least one authentication factor is stored in the local authentication model with a validity indication, indicating the time the authentication factor is valid for taking into account by the authentication engine, and authenticating a user by means of a match of the local authentication model with a set of user behavior indications retrieved by the user equipment through the authentication engine.

Abstract: A system configured to perform decision tasks carried out by a machine learning engine operates with a machine learning model, and includes a training component for improving the machine learning model, a device for carrying out decisions based on a set of input data, and an interaction interface for switching the machine learning model between training component and a device that includes a model attestation checker. The device performs acquiring input data, and ascertaining at least one machine learning model over the interaction interface. The model attestation checker performs checking if said machine learning model is trusted by a model attestation, and considering, for decision making, only those machine learning models that are trusted. The machine learning engine performs carrying out the decision task for input data by using a trusted machine learning model, and providing a result attestation for the decision output.

Abstract: The invention relates to a method for sending data to at least one device. According to the invention, a data sending control server sends to at least one data storage server at least one predetermined rule or a first request for sending data to at least one data processing server. The data storage server sends, based upon the at least one predetermined rule or the first request for sending data, data to the data processing server. The data sending control server sends to the data processing server a second request for sending to the device the data received or to be received by the data processing server. The data processing server sends, based upon the second request for sending the data, the received data to the at least one device.