Abstract: In general, the present disclosure describes techniques for detecting changes in demographic data of a customer based on energy consumption data of the customer. For example, a customer data management system receives energy consumption data of a customer and detects, based at least in part on the received energy consumption data of the customer, a change in demographic data associated with the customer. The customer data management system then outputs, based at least in part on the detecting, at least one demographic change report associated with the demographic data.

Abstract: 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.

Abstract: 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.

Abstract: A method for selecting an analysis procedure for a value series, including displaying a value series on a computer display monitor, receiving one or more sequences of user provided annotations, where the annotations overlay at least a sub-interval of the value series on the computer display monitor, using the sequences of user provided annotations to select an optimal value series analysis method from a set of value series analysis methods, where selecting an optimal value series analysis method includes determining parameter values for the optimal value series analysis method, and presenting the selected optimal value series analysis method and parameters, and the optimal reconstruction of the annotation sequences to the user.

Abstract: A method for accelerating time series data base queries includes segmenting an original time series of signal values into non-overlapping chunks, where a time-scale for each of the chunks is much less than the time scale of the entire time series, representing time series signal values in each chunk as a weighted superposition of atoms that are members of a shape dictionary to create a compressed time series, storing the original time series and the compressed time series into a database, determining whether a query is answerable using the compressed time series or the original time series, and whether answering the query using the compressed time series is faster. If answering the query is faster using the compressed representation, the query is executed on weight coefficients of the compressed time series to produce a query result, and the query result is translated back into an uncompressed representation.

Abstract: A predictive analysis generates a predictive model (Padj(Y|X)) based on two separate pieces of information, a set of original training data (Dorig), and a “true” distribution of indicators (Ptrue(X)). The predictive analysis begins by generating a base model distribution (Pgen(Y|X)) from the original training data set (Dorig) containing tuples (x,y) of indicators (x) and corresponding labels (y). Using the “true” distribution (Ptrue(X)) of indicators, a random data set (D?) of indicator records (x) is generated reflecting this “true” distribution (Ptrue(X)). Subsequently, the base model (Pgen(Y|X)) is applied to said random data set (D?), thus assigning a label (y) or a distribution of labels to each indicator record (x) in said random data set (D?) and generating an adjusted training set (Dadj). Finally, an adjusted predictive model (Padj(Y|X)) is trained based on said adjusted training set (Dadj).

Abstract: A method for accelerating time series data base queries includes segmenting an original time series of signal values into non-overlapping chunks, where a time-scale for each of the chunks is much less than the time scale of the entire time series, representing time series signal values in each chunk as a weighted superposition of atoms that are members of a shape dictionary to create a compressed time series, storing the original time series and the compressed time series into a database, determining whether a query is answerable using the compressed time series or the original time series, and whether answering the query using the compressed time series is faster. If answering the query is faster using the compressed representation, the query is executed on weight coefficients of the compressed time series to produce a query result, and the query result is translated back into an uncompressed representation.

Abstract: A method includes providing a columnar database comprising a plurality of columnar data structures associated with one column attribute; providing first data records having a plurality of first attribute-value pairs comprising counting information indicative of a number of first data records having the respective first attribute-value pair; providing mask data structures comprising one or more second attribute-value pairs; selecting second data records by intersecting the columnar data structures and the mask data structures; selecting one of the column attributes and one value contained in the column data structure associated with said selected column attribute as the destination attribute-value pair; creating one second rule for each first attribute-value pair; calculating, for each second rule, a co-occurrence-count between its respective source attribute-value pair and its destination attribute-value pair; and specifically selecting one or more of said second rules as the first rules in dependence on the

Abstract: Predictive analysis relating to nodes of a communication network is carried out by providing communication event information for a first set of nodes and a second set of nodes of the communication network, providing a set of attributes for the nodes of the first set, using the attributes and the communication event information for determining a set of groups among the first set of nodes, assigning each node of the second set to at least one group of the set of groups based at least on the communication event information available for the second group, the assigning resulting in membership information of the nodes of the second set, and deriving or applying a prediction model for the second set of nodes based on the communication event information for the second set and the membership information.

Abstract: A new data mining model (DMM) is created having at least one of the following characteristics: quality and complexity. The new DMM is handled as a candidate for storing in a storage device if a predefined criterion for the characteristics is met. The sum of the sizes of the new DMM and already stored DMMs is determined. In response to the sum falling below a storage limit, the new DMM is stored in the storage device. In response to the sum exceeding the storage limit, a decision is taken based on priorities of the DMMs which DMMs to store in the storage device.

Abstract: Computerized methods, data processing systems, and computer program products for storing of data mining models (DMMs) are provided. A new DMM is created having at least one of the following characteristics: quality and complexity. The new DMM is handled as a candidate for storing in a storage device if a predefined criterion for the characteristics is met. The sum of the sizes of the new DMM and already stored DMMs is determined. In response to the sum falling below a storage limit, the new DMM is stored in the storage device. In response to the sum exceeding the storage limit, a decision is taken based on priorities of the DMMs which DMMs to store in the storage device. The priorities depend at least on access frequencies of the DMMs. Upon a data mining request, a corresponding DMM is determined and a user is requested to confirm that data mining is to proceed if quality of the determined DMM does not fulfill a further predefined criterion.

Abstract: A new data mining model (DMM) is created having at least one of the following characteristics: quality and complexity. The new DMM is handled as a candidate for storing in a storage device if a predefined criterion for the characteristics is met. The sum of the sizes of the new DMM and already stored DMMs is determined. In response to the sum falling below a storage limit, the new DMM is stored in the storage device. In response to the sum exceeding the storage limit, a decision is taken based on priorities of the DMMs which DMMs to store in the storage device.

Abstract: A method includes providing a columnar database comprising a plurality of columnar data structures associated with one column attribute; providing first data records having a plurality of first attribute-value pairs comprising counting information indicative of a number of first data records having the respective first attribute-value pair; providing mask data structures comprising one or more second attribute-value pairs; selecting second data records by intersecting the columnar data structures and the mask data structures; selecting one of the column attributes and one value contained in the column data structure associated with said selected column attribute as the destination attribute-value pair; creating one second rule for each first attribute-value pair; calculating, for each second rule, a co-occurrence-count between its respective source attribute-value pair and its destination attribute-value pair; and specifically selecting one or more of said second rules as the first rules in dependence on the

Abstract: 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.

Abstract: A predictive analysis generates a predictive model (Padj(Y|X)) based on two separate pieces of information, a set of original training data (Dorig), and a “true” distribution of indicators (Ptrue(X)). The predictive analysis begins by generating a base model distribution (Pgen(Y|X)) from the original training data set (Dorig) containing tuples (x,y) of indicators (x) and corresponding labels (y). Using the “true” distribution (Ptrue(X)) of indicators, a random data set (D?) of indicator records (x) is generated reflecting this “true” distribution (Ptrue(X)). Subsequently, the base model (Pgen(Y|X)) is applied to said random data set (D?), thus assigning a label (y) or a distribution of labels to each indicator record (x) in said random data set (D?) and generating an adjusted training set (Dadj). Finally, an adjusted predictive model (Padj(Y|X)) is trained based on said adjusted training set (Dadj).

Abstract: The invention relates to a method for carrying out predictive analysis relating to nodes of a communication network. The method comprises the steps of providing communication event information for a first set of nodes and a second set of nodes of the communication network, providing a set of attributes for the nodes of the first set, using said attributes and said communication event information for determining a set of groups among the first set of nodes, assigning each node of the second set to at least one group of the set of groups based at least on the communication event information available for the second group, the assigning resulting in membership information of the nodes of the second set as well as deriving or applying a prediction model for the second set of nodes based on the communication event information for the second set and the membership information.

Abstract: A first data mining model and a second data mining model are compared. A first data mining model M1 represents results of a first data mining task on a first data set D1 and provides a set of first prediction values. A second data mining model M2 represents results of a second data mining task on a second data set D2 and provides a set of second prediction values. A relation R is determined between said sets of prediction values. For at least a first record of an input data set, a first and second probability distribution is created based on the first and second data mining models applied to the first record. A distance measure d is calculated for said first record using the first and second probability distributions and the relation. At least one region of interest is determined based on said distance measure d.

Abstract: Computerized methods, data processing systems, and computer program products for storing of data mining models (DMMs) are provided. A new DMM is created having at least one of the following characteristics: quality and complexity. The new DMM is handled as a candidate for storing in a storage device if a predefined criterion for the characteristics is met. The sum of the sizes of the new DMM and already stored DMMs is determined In response to the sum falling below a storage limit, the new DMM is stored in the storage device. In response to the sum exceeding the storage limit, a decision is taken based on priorities of the DMMs which DMMs to store in the storage device. The priorities depend at least on access frequencies of the DMMs. Upon a data mining request, a corresponding DMM is determined and a user is requested to confirm that data mining is to proceed if quality of the determined DMM does not fulfill a further predefined criterion.