Abstract: Methods, systems, and computer storage media for providing a dynamically weighted unobserved component model (“DW-UCM”) in a demand forecasting engine of a data analytics system. Dynamic weighting is performed based on a machine learning framework that includes tools, interfaces, and a library for developing improved machine learning models (e.g., dynamic demand forecasting models) of a dynamic weighting machine learning pipeline. In particular, the dynamic weighting machine learning pipeline can include a first module that is configured to predict if a segment (e.g., travel segment) under evaluation is open or closed (e.g., due to a restriction or rule), a second module that forecasts near-term recovery (e.g., approx. 0 - 4 weeks), and a third module that predicts longer term recovery.

Abstract: Methods and systems of using reinforcement learning to optimizing promotions. A promotion can be offered to a user using a reinforcement learning model with a sensitivity parameter, the reinforcement module estimating a time period during which the user will respond to the first information. The user's reaction to the promotion can be observed. The reinforcement learning model can be adapted based on the user's reaction.

Abstract: Methods and systems of using reinforcement learning for promotions. A first promotion is offered to a customer for a product and/or service. A first reward or penalty is determined, via a reinforcement machine learning model, based on the customer's reaction to the first promotion, wherein the reinforcement machine learning model is at a first state. Feedback to the reinforcement machine learning model is provided based on the first reward or penalty. A state of the reinforcement machine learning model is changed, based on the feedback, from the first state to a second state.

Abstract: Methods and Systems for using reinforcement learning to optimize promotions. A promotion can be offered to a customer for a prepaid calling card using a reinforcement learning model with a sensitivity parameter. The reinforcement learning model can estimate a time period during which the customer will purchase the prepaid calling card. The customer's reaction to the promotion can be observed. A reward or a penalty can be collected based on the customer's reaction. The reinforcement learning model can be adapted based on the reward or the penalty to optimize the timing of the promotion by estimating a new time period during which the customer will purchase the prepaid calling card. The reward proxy and/or the penalty proxy can comprise frequency of usage.

Abstract: Methods and systems for using reinforcement learning to optimizing promotions. A promotion for a product and/or service is offered to a customer using a reinforcement learning model. The customer's reaction is observed. A reward or a penalty is collected based on the customer's reaction. The reinforcement learning model is adapted based on the reward or penalty.

Abstract: Methods and systems for predicting are disclosed. Records for transactions can be stored, each record comprising an indication identifying items involved in a transaction and an indication of a time elapsed between the transaction and a previous transaction. The records for transactions can be analyzed to produce a probability that the items will be involved in a next transaction.

Abstract: Systems and methods for predicting are described herein. A record for each of a plurality of events associated with user transactions can be stored. A sequential plurality of the events can be analyzed using a unidirectional long short term memory (LSTM) and first and second dense neural network layers configured to receive output from the LSTM network.

Abstract: A processor circuit may perform processing associated with storing in a memory a record for each of a plurality of transactions, each record comprising an indication identifying at least one item, of a set of items, involved in the transaction and an indication of a time elapsed between the transaction and a previous transaction. The processor circuit may perform processing associated with analyzing a sequential plurality of the transactions using at least a long short term memory (LSTM) network to produce a probability that the at least one item will be involved in a next transaction for each at least one item, and a time estimate until the next transaction occurs. The processor circuit may perform processing associated with storing each probability and the time estimate in the memory as a prediction.

Abstract: Systems and methods for predicting transactions. A record is stored for each of a plurality of transactions, each record comprising an indication identifying each item from a set of items involved in the transaction and an indication of a time elapsed between the transaction and a previous transaction. A sequential plurality of the transactions is analyzed using a unidirectional long short term memory (LSTM) network to produce a probability that each item from the set of items will be involved in a next transaction, and a time estimate value for a future time when the next transaction will occur. The probability for each item and the time estimate value is stored in the memory as a prediction.

Abstract: A metamodel-based system for automatically generating reports is provided. Much of the work involved in creating report designs and reports is automated, allowing developers to produce a broad variety of report designs and reports in a fraction of the time required with standard report development, according to one illustrative embodiment. Metamodels for data and style templates may be created, that allow for the automated design and manipulation of report models. The report models may be used to generate reports that include the indicated data, arranged according to indicated data field properties, settings, and style templates. Changes made in the metamodels can automatically be reflected in reports designed according to those metamodels.

Abstract: A method of developing a report capable of providing information from a local database in a predefined structure to a remote application is discussed. The method includes defining an abstract data report structure. The abstract data report structure is configured to provide a report to the remote application having a plurality of data elements. The method further includes localizing at least one of the plurality of data elements by assigning it a remotely customizable field value; and exposing the abstract data report structure to the remote application.

Abstract: A system for use in rendering a report for a client application, client server application or multi-tier applications. The system includes a report definition generator that is configured to generate a report definition using metadata located in a metadata store. A database in the application configured to receive a query and to provide a data set. A design engine is configured to render the report based on the report definition, parameters and the data set.

Abstract: A method of developing a report capable of providing information from a local database in a predefined structure to a remote application is discussed. The method includes defining an abstract data report structure. The abstract data report structure is configured to provide a report to the remote application having a plurality of data elements. The method further includes localizing at least one of the plurality of data elements by assigning it a remotely customizable field value; and exposing the abstract data report structure to the remote application.

Abstract: A metamodel-based system for automatically generating reports is provided. Much of the work involved in creating report designs and reports is automated, allowing developers to produce a broad variety of report designs and reports in a fraction of the time required with standard report development, according to one illustrative embodiment. Metamodels for data and style templates may be created, that allow for the automated design and manipulation of report models. The report models may be used to generate reports that include the indicated data, arranged according to indicated data field properties, settings, and style templates. Changes made in the metamodels can automatically be reflected in reports designed according to those metamodels.

Abstract: A system for use in rendering a report for a client application, client server application or multi-tier applications. The system includes a report definition generator that is configured to generate a report definition using metadata located in a metadata store. A database in the application configured to receive a query and to provide a data set. A design engine is configured to render the report based on the report definition, parameters and the data set.

Abstract: Abstract metadata representing a multidimensional database expression and parameters in the multidimensional database expression are generated. Using the metadata, a class is generated that can be used to obtain an abstract data structure for the multidimensional database expression. The obtained data structure can then be translated into a desired multidimensional query language expression string that can be executed to retrieve data from a multidimensional database.