Abstract: A method includes generating an authoring representation of a machine learning pipeline based on a received input, where the authoring representation is configured to manage one or more machine learning operations. The method also includes receiving an indication of an operation to be performed on the authoring representation. The method further includes translating the authoring representation to an intermediate representation based on the operation and optimizing the intermediate representation. In addition, the method includes translating the intermediate representation to an execution representation that is understood by one or more machine learning executors.

Abstract: Apparatus, systems, and methods operate to partition a product feature set into a set of included features and a set of feature upselling candidates determined by a corresponding set of decision variables. The decision variables can be determined, in turn, by maximizing a profit function comprising a sum of first mathematical products (each product including profit, probability of sale, and a decision variable) over the product feature set, subject to the condition that a sum of second mathematical products (each product including feature utility and a decision variable) over the product feature set is less than or equal to the utility of the product feature set combined with a base product, minus the utility of a competing product. Operations may include designating an initial sales product configuration as a combination of a base product and the set of included features. Additional apparatus, systems, and methods are disclosed.

Abstract: Apparatus, systems, and methods operate to partition a product feature set into a set of included features and a set of feature upselling candidates determined by a corresponding set of decision variables. The decision variables can be determined, in turn, by maximizing a profit function comprising a sum of first mathematical products (each product including profit, probability of sale, and a decision variable) over the product feature set, subject to the condition that a sum of second mathematical products (each product including feature utility and a decision variable) over the product feature set is less than or equal to the utility of the product feature set combined with a base product, minus the utility of a competing product. Operations may include designating an initial sales product configuration as a combination of a base product and the set of included features. Additional apparatus, systems, and methods are disclosed.

Abstract: A method, as well as a computer program product that implements the method, is provided for user modeling, within a computer system, of a lifecycle of uniquely identified computer data objects that are members of a computer data object family. The method includes receiving a user-defined triggering event occurring in a software application system. The method also includes receiving user-defined permissible states for one or more object attributes that may be present for a uniquely identified computer data object that is a member of the data object family, as well as user-defined permissible transitions between the defined permissible states. The method also includes receiving a user-defined action to be performed on a uniquely identified computer data object that is a member of the computer data object family, when the user-defined triggering event occurs and causes a permissible state transition to occur.

Abstract: Techniques are provided for individual data objects in an enterprise computing system, where the individual data objects represent unique assets of a family of assets. The individual data objects are used in addition to business partner master data, and also in some cases in addition to generic product master data. The individual data objects are configurable such that attributes for an individual data object change over the lifetime of the unique asset to which the individual data object pertains. Techniques for modeling the lifecycle of an individual object family are described. Techniques also are provided for an integration framework that integrates a software application with an individual object that is used or altered by the software application.

Abstract: Techniques are provided for individual data objects in an enterprise computing system, where the individual data objects represent unique assets of a family of assets. The individual data objects are used in addition to business partner master data, and also in some cases in addition to generic product master data. The individual data objects are configurable such that attributes for an individual data object change over the lifetime of the unique asset to which the individual data object pertains. Techniques for modeling the lifecycle of an individual object family are described. Techniques also are provided for an integration framework that integrates a software application with an individual object that is used or altered by the software application.