Experian Information Solutions, Inc.

Systems and methods are described for machine learning-based generation of scorecards and boosted decision trees that facilitate explainable predictions. A scorecard machine learning model may be applied to historical records such that the model, for each of a number of variables, automatically generates (a) normal bins for normal values of the variable that fall within a valid range of values and (b) at least one special bin for special values of the variable that fall outside the valid range of values. Adjacent bins of the normal bins may be separated by a threshold value and each normal bin and each special bin may have an assigned score value. A risk assessment score may be generated based at least in part on the model identifying the assigned value score for each of the variables based on the normal or special bin to which each variable is assigned.

Embodiments of a data processing system comprises a first database including first records and a second database including second records. The data processing system may also comprise a processor configured to execute instructions to receive a first filter item comprising a first criterion and a second filter item comprising a second criterion and identify, in the first and second records, at least one record that meets the first criterion of the first filter item. The instructions may update the at least one record to include a unique identifier and generate a third database including third records by identifying, in the first and second records, records that include the unique identifier. The instructions may cause a comparison of a rubric of historical linkage features with the third records to identify an element of historical linkage between the at least one record and the third records and generate an event notification including the element of historical linkage.

Systems and methods are described for identifying a subset of interest from a general population and for monitoring a database of daily activity logs associated with the general population in order to identify database entries indicative of an occurrence of a pre-defined trigger event that is associated with a member of the subset of interest. In particular, systems and methods are described that allow a massive database of daily activity logs to be monitored to identify trigger events that have occurred within the past twenty-four hours or other very recent time period. Embodiments are described that may be advantageously used by a provider of credit-related products and/or services who wishes to accurately target prospective customers, identified by the system, based on occurrence of a trigger event, as being in a decision-making phase of credit shopping, for purposes of making a timely and targeted offering relevant to the customers' current activities.

Embodiments of a data processing system comprises a first database including first records and a second database including second records. The data processing system may also comprise a processor configured to execute instructions to receive a first filter item comprising a first criterion and a second filter item comprising a second criterion and identify, in the first and second records, at least one record that meets the first criterion of the first filter item. The instructions may update the at least one record to include a unique identifier and generate a third database including third records by identifying, in the first and second records, records that include the unique identifier. The instructions may cause a comparison of a rubric of historical linkage features with the third records to identify an element of historical linkage between the at least one record and the third records and generate an event notification including the element of historical linkage.

The features relate to artificial intelligence directed compression of user event data based on complex analysis of user event data including latent feature detection and clustering. Further features are described for reducing the size of data transmitted during event processing data flows and devices such as card readers or point of sale systems. Machine learning features for dynamically determining an optimal compression as well as identifying targeted users and providing content to the targeted users based on the compressed data are also included.