Abstract: A domain fact verification system is described having a computer programmed with a model trained using a process of data distillation and model distillation to improve model learning of the underlying semantics of a dataset rather than relying on statistical and lexical nuances in a domain-specific dataset. The computer thus programmed can accurately perform fact verification across multiple domains without the labor-intensive process of encoding a dataset of human-annotated, domain-specific information for each domain. Moreover, by combining data distillation with model distillation techniques, which may be seen as an inverse of well-established ensemble strategies (which train individual models separately and applies them jointly) the present domain transferable fact verification system scales better at inference time due to its reliance on a single trained model.

Abstract: The technique relates to a method and apparatus for dynamic destruction of data in a remote data storage platform. This involves receiving a first set of data records with random values and then updating the random values associated with the subsets of the first set of data records into actual values through one or more update operations. After sequential steps of updates, a second set of data records with actual values are obtained. After working of this data for a period of time the owner of the data again begins updating the actual values associated with the subset of the second set of data records into random values. When all the data records are updated to random values the owner of the data successfully ends the Service Level Agreement period.

Abstract: The technique relates to a method and apparatus for dynamic destruction of data in a remote data storage platform. This involves receiving a first set of data records with random values and then updating the random values associated with the subsets of the first set of data records into actual values through one or more update operations. After sequential steps of updates, a second set of data records with actual values are obtained. After working of this data for a period of time the owner of the data again begins updating the actual values associated with the subset of the second set of data records into random values. When all the data records are updated to random values the owner of the data successfully ends the Service Level Agreement period.

Abstract: A system and method for presenting on-demand masking of data as a software service in a distributed environment is provided. An application hosted on a computing device receives request for access to application data from a user. Credentials of the user are first validated in order to determine whether the user is authorized to access the requested application data. For an authorized user, a category of the user is determined to ascertain whether the user is privileged to obtain full access. In case the user is a privileged user, unmasked application data is fetched from a database utility and provided to the user. In case the user is not a privileged user, application data access request is transferred to a data masking service. Application data is fetched from database utility, masked based on pre-defined masking rules and provided to the user.

Abstract: Embodiments of a system and method to perform a secure deletion of a set of data from a remote cloud computing system are described. As described, in some embodiments, a user of a cloud computing service that provides data storage may securely delete his stored set of data by acquiring elevated access privileges to the stored set of data, designating at least one most significant bit in at least one data block therein as a sentinel and recording its value and position, updating the value of the sentinels and thereby rendering the data unusable, and verifying the success of the operation by checking the new value of the sentinels against the original value. In some embodiments, the verification process may be repeated in order to ensure that the data has been rendered useless across all nodes of the remote cloud platform.

Abstract: Embodiments of a system and method to perform a secure deletion of a set of data from a remote cloud computing system are described. As described, in some embodiments, a user of a cloud computing service that provides data storage may securely delete his stored set of data by acquiring elevated access privileges to the stored set of data, designating at least one most significant bit in at least one data block therein as a sentinel and recording its value and position, updating the value of the sentinels and thereby rendering the data unusable, and verifying the success of the operation by checking the new value of the sentinels against the original value. In some embodiments, the verification process may be repeated in order to ensure that the data has been rendered useless across all nodes of the remote cloud platform.

Abstract: A system and method for presenting on-demand masking of data as a software service in a distributed environment is provided. An application hosted on a computing device receives request for access to application data from a user. Credentials of the user are first validated in order to determine whether the user is authorized to access the requested application data. For an authorized user, a category of the user is determined to ascertain whether the user is privileged to obtain full access. In case the user is a privileged user, unmasked application data is fetched from a database utility and provided to the user. In case the user is not a privileged user, application data access request is transferred to a data masking service. Application data is fetched from database utility, masked based on pre-defined masking rules and provided to the user.