Abstract: One embodiment provides automatically learning shared resource environment solution design rules from a collection of requirement-solution pairs including obtaining requirement-solution pairs for a shared resource environment from a data store. A processor iteratively generates a candidate design rule set from each requirement-solution pair. Each generating iteration uses an input including the candidate design rule set output from a previous generating iteration. Evidence scores of each candidate design rule are calculated and candidate design rules having higher evidence score than an evidence score threshold are retained to obtain a learned design rule set. Candidate rules of a next iteration are constructed based on an addition of new attributes to rules of the learned design rule set.

Abstract: One embodiment provides for generating a cognitive executable process graph including obtaining, by a processor, a hybrid process knowledge graph generated based process fragments and a set of actionable statements and business constraints. The hybrid process knowledge graph including different node types. The hybrid knowledge graph is traversed from a root of a process through each task in the hybrid process knowledge graph to obtain an action and metadata for each task node. Based on the action and metadata, at least one statement in an equivalent executable code block is created to represent the action. A cognitive executable process graph is generated based on at least one executable code block.

Abstract: One embodiment provides model-driven and automated generation of information technology (IT) solutions including obtaining a set of business and technical requirements for IT infrastructure and applications. A client business and technical requirement model is generated based on generic model constructs and extending with constructs specific to capturing client requirements. A draft IT solution is generated using an automated model-driven process to generate the draft IT solution configuration for client requirements for a target shared resource environment offering. The generated draft IT solution is translated into a language of a constraint satisfaction engine that propagates values of chosen attributes in the draft solution to identify valid values for unset attributes, and identifies conflicts. An IT solutions interface is generated based on auto-population of verified attribute results.

Abstract: One embodiment provides a method comprising extracting natural language content from a piece of communication for a user, generating a representation of the piece of communication based on the natural language content extracted, and utilizing a global deep learning model and a personalized learning model for the user to assign a priority label to the piece of communication based on the representation and user behavioral information associated with recent conversations of the user. Another embodiment provides a method comprising, for each piece of communication of a set of multiple pieces of communication for multiple users, extracting natural language content from the piece communication and generating a representation of the piece of communication based on the natural language extracted, and training a deep learning neural network to predict a degree of priority of a subsequent piece of communication based on each representation generated.

Abstract: One embodiment provides a method comprising mapping project attributes for past projects to a first parameter set associated with a first model that models distribution of event types of project events, and a second parameter set associated with a second model that models distribution of the time intervals of project events. Specifically, machine learning is applied to a set of historical data for the past projects to obtain a first and a second set of learned weights. The method further comprises predicting information relating to a next project event for an ongoing project by generating a first probability distribution for a set of possible event types for the next project event utilizing the first model, and, for each possible event type, generating a corresponding probability distribution for time intervals of occurrence of the possible event type utilizing the first model and the second model in a pipelined fashion.

Abstract: One embodiment provides automatically learning shared resource environment solution design rules from a collection of requirement-solution pairs including obtaining requirement-solution pairs for a shared resource environment from a data store. A processor iteratively generates a candidate design rule set from each requirement-solution pair. Each generating iteration uses an input including the candidate design rule set output from a previous generating iteration. Evidence scores of each candidate design rule are calculated and candidate design rules having higher evidence score than an evidence score threshold are retained to obtain a learned design rule set. Candidate rules of a next iteration are constructed based on an addition of new attributes to rules of the learned design rule set.

Abstract: One embodiment provides automatically learning shared resource environment solution design rules from a collection of requirement-solution pairs including obtaining requirement-solution pairs. A processor iteratively generates a candidate design rule set from each requirement-solution pair. Candidate design rules from the candidate rule set are filtered to obtain a learned design rule set. The learned design rule set is optimized based on merging design rules.

Abstract: One embodiment provides for continuously adaptive business process management definition and execution including generating a continuously adaptive business process model and execution environment. New goals are discovered. Entity information is extracted from input documents. A model knowledge graph is generated that includes a first parse-tree for process fragments using the discovered new goals and the extracted entity information.

Abstract: One embodiment provides discovering knowledge rich and executable business process models from unstructured information including obtaining, by a processor, unstructured data source information describing business processes. Based on the unstructured data source information, an executable specification of described business processes and a corresponding amendable textual specification are generated. Business process models are generated using a process knowledge graph based on the executable specification.

Abstract: One embodiment provides for continuously adaptive business process management definition and execution including generating a continuously adaptive business process model and execution environment. New goals are discovered. Entity information is extracted from input documents. A model knowledge graph is generated that includes a first parse-tree for process fragments using the discovered new goals and the extracted entity information.

Abstract: One embodiment provides for continuously adaptive business process management definition and execution including obtaining, by a processor, business process models and a process runtime environment. A business process model is discovered. Business rules that support decision making for the business process model are discovered. A process plan is defined in view of the business rules for achieving a predetermined goal. A next action in the process plan is determined based on the business rules in a current process portion and providing a recommendation for acting on the next action. The next action is executed based on the recommendation. A change in a world effect status is determined after executing the next action. The process plan is updated. A continuously adaptive business process model and execution environment are generated.

Abstract: One embodiment provides a method comprising extracting natural language content from a piece of communication for a user, generating a representation of the piece of communication based on the natural language content extracted, and utilizing a global deep learning model and a personalized learning model for the user to assign a priority label to the piece of communication based on the representation and user behavioral information associated with recent conversations of the user. Another embodiment provides a method comprising, for each piece of communication of a set of multiple pieces of communication for multiple users, extracting natural language content from the piece communication and generating a representation of the piece of communication based on the natural language extracted, and training a deep learning neural network to predict a degree of priority of a subsequent piece of communication based on each representation generated.

Abstract: One embodiment provides for predicting and planning of staffing needs for services including obtaining data from an opportunity pipeline. The data including current and historical project information, offerings information included in each opportunity and current and historical staffing information. An optimization model is generated to provide a threshold for deals predicted to be won. A threshold of win score for deals to be considered as predicted to be won is optimized. Opportunities to be won are predicted including: executing a win prediction model for current opportunities in the opportunity pipeline, filtering deals with scores less than the win score threshold, processing a deal progress monitoring model for each remaining deal to predict a future event and related timeline, and simulating progress of each deal by updating each deal with a predicted event until an end of a simulation time window.

Abstract: According to an aspect, document-based requirement identification and extraction includes parsing a set of documents and identifying relationships among parsed components of the documents and applying the parsed components and identified relationships to a meta-model that defines requirements. The requirements include a statement expressing a need and/or responsibility. A further aspect includes identifying candidate requirements and their candidate topics from results of the applying. For each of the identified candidate topics, a feature vector is built from the corresponding candidate requirements. A further aspect includes training the meta-model with the feature vectors, validating the meta-model, and classifying output of the validating to identify a subset of the candidate requirements, and corresponding topics expressed in the set of documents.

Abstract: One embodiment provides discovering knowledge rich and executable business process models from unstructured information including obtaining, by a processor, unstructured data source information describing business processes. Based on the unstructured data source information, an executable specification of described business processes and a corresponding amendable textual specification are generated. Business process models are generated using a process knowledge graph based on the executable specification.

Abstract: One embodiment provides for generating a cognitive executable process graph including obtaining, by a processor, a hybrid process knowledge graph generated based process fragments and a set of actionable statements and business constraints. The hybrid process knowledge graph including different node types. The hybrid knowledge graph is traversed from a root of a process through each task in the hybrid process knowledge graph to obtain an action and metadata for each task node. Based on the action and metadata, at least one statement in an equivalent executable code block is created to represent the action. A cognitive executable process graph is generated based on at least one executable code block.

Abstract: One embodiment provides for continuously adaptive business process management definition and execution including obtaining, by a processor, business process models and a process runtime environment. A business process model is discovered. Business rules that support decision making for the business process model are discovered. A process plan is defined in view of the business rules for achieving a predetermined goal. A next action in the process plan is determined based on the business rules in a current process portion and providing a recommendation for acting on the next action. The next action is executed based on the recommendation. A change in a world effect status is determined after executing the next action. The process plan is updated. A continuously adaptive business process model and execution environment are generated.

Abstract: One embodiment provides model-driven and automated generation of information technology (IT) solutions including obtaining a set of business and technical requirements for IT infrastructure and applications. A client business and technical requirement model is generated based on generic model constructs and extending with constructs specific to capturing client requirements. A draft IT solution is generated using an automated model-driven process to generate the draft IT solution configuration for client requirements for a target shared resource environment offering. The generated draft IT solution is translated into a language of a constraint satisfaction engine that propagates values of chosen attributes in the draft solution to identify valid values for unset attributes, and identifies conflicts. An IT solutions interface is generated based on auto-population of verified attribute results.

Abstract: One embodiment provides automatically learning shared resource environment solution design rules from a collection of requirement-solution pairs including obtaining requirement-solution pairs. A processor iteratively generates a candidate design rule set from each requirement-solution pair. Candidate design rules from the candidate rule set are filtered to obtain a learned design rule set. The learned design rule set is optimized based on merging design rules.

Abstract: A computer-implemented method includes learning one or more historical curves based on historical data that describes two or more historical pipelines of sales contracts. Each of the one or more historical curves is based on a corresponding historical pipeline of the two or more historical pipelines. Two or more similarity indices are generated, where each similarity index corresponds to a historical pipeline of the two or more historical pipelines, and each similarity index is based on similarity between the corresponding historical pipeline and a target pipeline for which a prediction is sought. A first curve is fit, by a computer processor, to the target pipeline, where the target pipeline has unknown data, and the first curve is based on the two or more similarity indices. A pipeline value of the target pipeline is predicted based on the first curve.