摘要:
An implementation of NMF functionality integrated into a relational database management system provides the capability to apply NMF to relational datasets and to sparse datasets. A database management system comprises a multi-dimensional data table operable to store data and a processing unit operable to perform non-negative matrix factorization on data stored in the multi-dimensional data table and to generate a plurality of data tables, each data table being smaller than the multi-dimensional data table and having reduced dimensionality relative to the multi-dimensional data table. The multi-dimensional data table may be a relational data table.
摘要:
The present invention relates to progress notification systems, computer program products and methods of operation thereof, that reports processing progress of data mining operations at regular periodic intervals. The system comprises: an input/output interface for exchanging information with a network; a memory for storing updated progress objects associated with the data mining operation as a set of data mining algorithms progress in processing; and a processor coupled to the input/output interface and the memory, the processor for performing the data mining operation, the data mining operation implementing the set of data mining algorithms; and generating a notification object for the data mining operation at a pre-determined interval, the notification object based on the progress objects at each of the pre-determined intervals.
摘要:
An implementation of NMF functionality integrated into a relational database management system provides the capability to apply NMF to relational datasets and to sparse datasets. A database management system comprises a multi-dimensional data table operable to store data and a processing unit operable to perform non-negative matrix factorization on data stored in the multi-dimensional data table and to generate a plurality of data tables, each data table being smaller than the multi-dimensional data table and having reduced dimensionality relative to the multi-dimensional data table. The multi-dimensional data table may be a relational data table.
摘要:
A system, method, and computer program product that uses attribute importance (AI) to reduce the time and computation resources required to build data mining models, and which provides a corresponding reduction in the cost of data mining. Attribute importance (AI) involves a process of choosing a subset of the original predictive attributes by eliminating redundant, irrelevant or uninformative ones and identifying those predictor attributes that may be most helpful in making predictions. A new algorithm Predictor Variance is proposed and a method of selecting predictive attributes for a data mining model comprises the steps of receiving a dataset having a plurality of predictor attributes, for each predictor attribute, determining a predictive quality of the predictor attribute, selecting at least one predictor attribute based on the determined predictive quality of the predictor attribute, and building a data mining model including only the selected at least one predictor attribute.
摘要:
A system, method, and computer program product provides a useful measure of the accuracy of a Naïve Bayes predictive model and reduced computational expense relative to conventional techniques. A method for measuring accuracy of a Naïve Bayes predictive model comprises the steps of receiving a training dataset comprising a plurality of rows of data, building a Naïve Bayes predictive model using the training dataset, for each of at least a portion of the plurality of rows of data in the training dataset incrementally untraining the Naïve Bayes predictive model using the row of data and determining an accuracy of the incrementally untrained Naïve Bayes predictive model, and determining an aggregate accuracy of the Naïve Bayes predictive model.
摘要:
The notion of checkpointing model building is introduced to safeguard against the loss of computational results due to abnormal termination of long running computation algorithms. A checkpoint manager initiates a checkpoint, wherein initiation occurs by various scenarios, including dynamically, automatically and manually. A computation module executes the checkpoint on an in progress data computation. The checkpoint manager also monitors the execution of the checkpoint for abnormal termination. Upon a determination of abnormal checkpoint termination, the inability to resume model building from a checkpoint, the checkpoint manger either resumes model build from the checkpoint or aborts the model build.