Abstract: Systems and methods for composite storage of data across tiers of a multi-tenant storage device are provided. In embodiments, a method includes determining attributes associated with each query in a set of queries; determining whether each of the attributes meet respective predetermined threshold values for a first storage tier, a second storage tier and a third storage tier based on statistical data generated for the attributes; updating storage rules for at least one of the first storage tier, the second storage tier and the third storage tier based on the determining; receiving data from the tenant for storage; dividing the data into at least a first portion and a second portion based on the storage rules; and storing the at least the first portion and the second portion on separate ones of the first storage tier, the second storage tier and the third storage tier based on the storage rules.

Abstract: A computer-implemented method for transferring data is provided. In an illustrative embodiment, the method includes retrieving, by a computer, an original dataset to be sent from a sender to a receiver. The method also includes generating, by the computer, a model based on at least a subset of the original dataset. The model generates a predicted dataset. The model is selected from a plurality of model types based on data complexity of the original dataset and a desired level of approximation of the predicted dataset to the original dataset. The method also includes transferring, by the computer, the model to the receiver. The receiver uses the model to generate the predicted dataset, wherein the predicted dataset matches the original dataset to a selected degree of approximation. Transfer of the model is quicker than transfer of the original dataset.

Abstract: A computer-implemented method for determining an evaluation metric value for a multi-class classifier with unknown labels may be provided. The method comprises providing a trained classifier system for classifying input data vectors resulting in a plurality of result label values and respective probability vectors, predicting, based on received input data vectors, result label values and respective probability vectors, and receiving feedback data comprising respective feedback label values. The method comprises also mapping each of the feedback label values to a respective highest probability dimension value of the related probability vector resulting in an index map, and determining, based on the index map, an evaluation metric value for the quality of the trained classifier system.

Abstract: A method for controlling exposure of sensitive data though a logging system is provided. The method comprises: upon receiving a request by the request handler, determining sensitive data as part of the request by applying a rule, converting the data into a transformed format, and registering the data together with a related data field label with a log handler. Then, upon receiving by the log handler a log entry, converting each expression of the log entry into the transformed format, and comparing each transformed expression with each of the sensitive data in the transformed format. Upon determining a match of one of the transformed expressions with one of the sensitive data in the transformed format, the method comprises issuing an alert indicating that the log entry comprises sensitive data.

Abstract: A computer-implemented method for transferring data is provided. In an illustrative embodiment, the method includes retrieving, by a computer, an original dataset to be sent from a sender to a receiver. The method also includes generating, by the computer, a model based on at least a subset of the original dataset. The model generates a predicted dataset. The model is selected from a plurality of model types based on data complexity of the original dataset and a desired level of approximation of the predicted dataset to the original dataset. The method also includes transferring, by the computer, the model to the receiver. The receiver uses the model to generate the predicted dataset, wherein the predicted dataset matches the original dataset to a selected degree of approximation. Transfer of the model is quicker than transfer of the original dataset.

Abstract: A method for controlling exposure of sensitive data though a logging system is provided. The method comprises: upon receiving a request by the request handler, determining sensitive data as part of the request by applying a rule, converting the data into a transformed format, and registering the data together with a related data field label with a log handler. Then, upon receiving by the log handler a log entry, converting each expression of the log entry into the transformed format, and comparing each transformed expression with each of the sensitive data in the transformed format. Upon determining a match of one of the transformed expressions with one of the sensitive data in the transformed format, the method comprises issuing an alert indicating that the log entry comprises sensitive data.

Abstract: Systems and methods for composite storage of data across tiers of a multi-tenant storage device are provided. In embodiments, a method includes determining attributes associated with each query in a set of queries; determining whether each of the attributes meet respective predetermined threshold values for a first storage tier, a second storage tier and a third storage tier based on statistical data generated for the attributes; updating storage rules for at least one of the first storage tier, the second storage tier and the third storage tier based on the determining; receiving data from the tenant for storage; dividing the data into at least a first portion and a second portion based on the storage rules; and storing the at least the first portion and the second portion on separate ones of the first storage tier, the second storage tier and the third storage tier based on the storage rules.

Abstract: A computer-implemented method for determining an evaluation metric value for a multi-class classifier with unknown labels may be provided. The method comprises providing a trained classifier system for classifying input data vectors resulting in a plurality of result label values and respective probability vectors, predicting, based on received input data vectors, result label values and respective probability vectors, and receiving feedback data comprising respective feedback label values. The method comprises also mapping each of the feedback label values to a respective highest probability dimension value of the related probability vector resulting in an index map, and determining, based on the index map, an evaluation metric value for the quality of the trained classifier system.

Abstract: In a method for machine learning model training, the method includes one or more processors receiving a trained original machine learning model, including related parameters and a set of training data with which the machine learning model has been trained. The method further includes one or more processors determining an original quality evaluation value for the trained original machine learning model using a first set of feedback data. The method further includes one or more processors, in response to determining that the quality evaluation value is below a quality threshold value, triggering a retraining process for the original machine learning model, the retraining process comprising a first retraining phase for a first machine learning model and a second retraining phase for a second machine learning model.

Abstract: A method for controlling exposure of sensitive data though a logging system is provided. The method comprises: upon receiving a request by the request handler, determining sensitive data as part of the request by applying a rule, converting the data into a transformed format, and registering the data together with a related data field label with a log handler. Then, upon receiving by the log handler a log entry, converting each expression of the log entry into the transformed format, and comparing each transformed expression with each of the sensitive data in the transformed format. Upon determining a match of one of the transformed expressions with one of the sensitive data in the transformed format, the method comprises issuing an alert indicating that the log entry comprises sensitive data.

Abstract: A method for controlling exposure of sensitive data though a logging system is provided. The method comprises: upon receiving a request by the request handler, determining sensitive data as part of the request by applying a rule, converting the data into a transformed format, and registering the data together with a related data field label with a log handler. Then, upon receiving by the log handler a log entry, converting each expression of the log entry into the transformed format, and comparing each transformed expression with each of the sensitive data in the transformed format. Upon determining a match of one of the transformed expressions with one of the sensitive data in the transformed format, the method comprises issuing an alert indicating that the log entry comprises sensitive data.

Abstract: An approach is provided to registering a database at a database manager in a manner that defines software bundles. Registering the database involves receiving database metadata that corresponds to the database which is stored in a metadata data store that is maintained by the database manager. A software product associated with the registered database is processed by receiving software product metadata corresponding to the software product and this software product metadata is also stored in the metadata data store maintained by the database manager.