International Business Machines Corporation

Michael Wurst

Pascal Pompey

Aris Gkoulalas-Divanis

Olivier Verscheure

Christoph Lingenfelder

Martin Oberhofer

Stefano Braghin

Mathieu Sinn

Patrick Dantressangle

Francesco Fusco

Jakub Marecek

Dimitrios Mavroeidis

Carlos A Alzate Perez

Eberhard Hechler

Alexander Lang

Christoph H Sieb

Bernhard Mitschang

Ruben Pulido De Los Reyes

Torsten Steinbach

James L Finnie

Quasi-identifiers (QIDs) are detected in a dataset using a set of computing tasks. The dataset has a plurality of records and a set of attributes. An index is generated for the dataset. The index has an indicator for each attribute value of each record in the dataset. Each indicator specifies all the records in the dataset having the same value for the attribute. Each task is assigned an attribute combination and a subset of the plurality of records in the dataset and is passed to a thread for execution on computing resources. The executing task inspects the set of records specified by the index indicator for each attribute value in the attribute combination to produce a result. The result of at least one task identifies a unique record for the associated attribute combination. The attribute combination producing the unique record is a QID.

Detecting quasi-identifiers in datasets

The cost of data-mining is estimated where data-mining services are delivered via a distributed computing system environment. System requirements are estimated for a particular data-mining task for an input data set having specified properties. Estimating system requirements includes applying a partial learning tool to operate on sample data from the input data set.

Estimating the cost of data-mining services

In an approach for adaptive sampling of smart meter data, a computer retrieves one or more balancing constraints associated with one or more smart meter sensors. The computer retrieves meter sensor data from the one or more smart meter sensors according to the one or more balancing constraints. The computer determines a subsample of the meter sensor data based, at least in part, on one or more similar consumption patterns of meter sensor data, and then transmits the subsample of the meter sensor data to an optimization engine for use in solving an optimization problem.

Adaptive sampling of smart meter data

Embodiments relate to managing resource consumption in a computing system. An aspect includes providing a resource policy by defining a plurality of threshold values relating to the resource consumption, wherein the resources are consumed by a plurality of user-defined functions performing tasks for a database management system, wherein the user-defined functions are executed by a plurality of processes external to the database management system. Another aspect includes performing an action, as defined by the resource policy, on at least one of the user-defined functions.

Managing resource consumption in a computing system

At least one user table in a relational database management system (RDBMS) using a first operator within a structured query language (SQL) command is identified. The first operator within the SQL command is utilized to transfer one or more data items from the at least one user table to a data array within the RDBMS. The data array is processed within the RDBMS, and one or more output values are generated based on the processing.

Data management in relational databases

In an approach for adaptive sampling of smart meter data, a computer retrieves one or more balancing constraints associated with one or more smart meter sensors. The computer retrieves meter sensor data from the one or more smart meter sensors according to the one or more balancing constraints. The computer determines a subsample of the meter sensor data, and then transmits the subsample of the meter sensor data to an optimization engine for use in solving an optimization problem.

As information becomes more accessible to the public, the ability to predict and estimate sensitive data from the data already available to the general public becomes easier. The existing privacy-preserving data mining approaches only consider the information the user is querying and do not consider the information the user already has, and how the user can use that information in combination with the query information to create sensitive data that the user should not have access to. Some embodiments of the present invention provide a query analysis (QA) program that solves the aforementioned problem by taking into account data that a user may already have, whether it is private data or data that is available to the public, and then using that data, along with the data that would be returned in the query, to determine if sensitive data could be recreated.

Restricting sensitive query results in information management platforms

Accessing data in a database includes receiving, from a first user, a first query for a dataset stored in a database. A first set of patterns is provided in the dataset. For each pattern in the first set of patterns, a significance value is provided in response to the received first query. A set of tags is provided for flagging a pattern of the first set of patterns, the set of tags indicating at least two data categories describing the pattern. Input information received from the first user indicates tags of at least a first subset of patterns of the first set of patterns, wherein each tag of the tags is selected from the set of tags. The significance values of the first subset of patterns are adjusted based on the tags.

Mining patterns in a dataset

Techniques for iterative feature extraction using domain knowledge are provided. In one aspect, a method for feature extraction is provided. The method includes the following steps. At least one query to predict at least one future value of a given value series based on a statistical model is received. At least two predictions of the future value are produced fulfilling at least the properties of 1) each being as probable as possible given the statistical model and 2) being mutually divert (in terms of numerical distance measure). A user is queried to select one of the predictions. The user may be queried for textual annotations for the predictions. The annotations may be used to identify additional covariates to create an extended set of covariates. The extended set of covariates may be used to improve the accuracy of the statistical model.

Growing community of inventors

Stuttgart, Germany

Michael Wurst