Abstract: Mechanisms are provided for forecasting information technology (IT) and environmental impacts on key performance indicators (KPIs). Machine learning (ML) computer model(s) are trained on historical data representing IT events and KPIs of organizational processes (OPs). The ML computer model(s) forecast IT events given KPIs, or KPI impact given IT events. Correlation graph data structure(s) are generated that map at least one of IT events to IT computing resources, or KPI impacts to OPs. The trained ML computer model(s) process input data to generate a forecast output that specifies at least one of a forecasted IT event or a KPI impact. The forecasted output is correlated with at least one of IT computing resource(s) or OP(s), at least by applying the correlation graph data structure(s) to the forecast output to generate a correlation output. A remedial action recommendation is generated based on the forecast output and correlation output.

Abstract: Methods, systems and computer program products for automated learning are provided herein. A computer-implemented method includes authenticating a plurality of users for an automated learning session, wherein the plurality of users correspond to at least one device, and providing the automated learning session for the plurality of users. Providing the automated learning session comprises analyzing a plurality of learning models corresponding to one or more of the plurality of users, determining, based on the analysis, one or more activities to be performed by the plurality of users during the automated learning session, and executing the one or more activities on at least one device.

Abstract: Provided is a computer-implemented method, system, and computer program product for process mining asynchronous support conversations using attributed directly follows graphing. A processor may collect a plurality of conversation threads from an asynchronous data stream. The processor may label each utterance of a plurality of utterances from the plurality of conversation threads with an event label. The processor may analyze the event label for each utterance of the plurality of utterances. The processor may generate, based on the analyzing of the event label for each utterance, an attributed directly follows graph (DFG).

Abstract: Methods, systems, and computer program products for generating organizational goal-oriented and process conformant recommendation models using artificial intelligence techniques are provided herein.

Abstract: Methods, systems, and computer program products for content marshaling using biometric data are provided herein. A computer-implemented method includes estimating a cognitive state of a user based at least in part on analyzing one or more sets of biometric data pertaining to the user; dynamically compiling multi-modal content for the user based at least in part on (i) the estimated cognitive state of the user, (ii) one or more user parameters, and (iii) content availability information; and outputting the compiled content to the user via one or more devices.

Abstract: One embodiment provides a method, including: receiving, in an environment for developing a conversational tutor, domain content corresponding to a domain of the conversational tutor, wherein the conversational tutor comprises a conversational agent designed to assist students in learning a domain; receiving, at the environment, tutoring content developed by a user, wherein the tutoring content comprises a tutoring script comprising questions to be presented to the student during a tutoring session with the conversational tutor to assist the student in learning the domain; evaluating, within the environment, the tutoring content against the domain content, wherein the evaluating comprises (i) identifying missing tutoring content, (ii) identifying the difficulty of the tutoring content, and (iii) identifying question dependencies; and providing, at the environment, at least one recommendation, identified from results of the evaluation, for adjusting the tutoring content.

Abstract: One embodiment provides a method, including: receiving a plurality of transcripts, each transcript corresponding to an interaction between a student and a tutoring agent; identifying conversational patterns within the plurality of transcripts, the identified conversational patterns indicating a level of understanding regarding the interaction between the tutoring agent and the student, wherein the identifying comprises (i) identifying an exchange between the tutoring agent and the student, the exchange comprising a conversational turn by each of the tutoring agent and the student and (ii) classifying each of the conversational turns into a response type; determining, based upon the identified conversational patterns, a possible improvement to the tutoring agent, wherein the possible improvement is determined based upon the classified response types of each of (i) the tutoring agent and (ii) the student; and notifying a developer of the tutoring agent of the identified possible improvement.

Abstract: Methods, systems and computer program products for automated learning are provided herein. A computer-implemented method includes authenticating a plurality of users for an automated learning session, wherein the plurality of users correspond to at least one device, and providing the automated learning session for the plurality of users. Providing the automated learning session comprises analyzing a plurality of learning models corresponding to one or more of the plurality of users, determining, based on the analysis, one or more activities to be performed by the plurality of users during the automated learning session, and executing the one or more activities on at least one device.

Abstract: Methods, systems, and computer program products for content marshaling using biometric data are provided herein. A computer-implemented method includes estimating a cognitive state of a user based at least in part on analyzing one or more sets of biometric data pertaining to the user; dynamically compiling multi-modal content for the user based at least in part on (i) the estimated cognitive state of the user, (ii) one or more user parameters, and (iii) content availability information; and outputting the compiled content to the user via one or more devices.

Abstract: One embodiment provides a method, including: receiving, in an environment for developing a conversational tutor, domain content corresponding to a domain of the conversational tutor, wherein the conversational tutor comprises a conversational agent designed to assist students in learning a domain; receiving, at the environment, tutoring content developed by a user, wherein the tutoring content comprises a tutoring script comprising questions to be presented to the student during a tutoring session with the conversational tutor to assist the student in learning the domain; evaluating, within the environment, the tutoring content against the domain content, wherein the evaluating comprises (i) identifying missing tutoring content, (ii) identifying the difficulty of the tutoring content, and (iii) identifying question dependencies; and providing, at the environment, at least one recommendation, identified from results of the evaluation, for adjusting the tutoring content.

Abstract: One embodiment provides a method, including: receiving a plurality of transcripts, each transcript corresponding to an interaction between a student and a tutoring agent; identifying conversational patterns within the plurality of transcripts, the identified conversational patterns indicating a level of understanding regarding the interaction between the tutoring agent and the student, wherein the identifying comprises (i) identifying an exchange between the tutoring agent and the student, the exchange comprising a conversational turn by each of the tutoring agent and the student and (ii) classifying each of the conversational turns into a response type; determining, based upon the identified conversational patterns, a possible improvement to the tutoring agent, wherein the possible improvement is determined based upon the classified response types of each of (i) the tutoring agent and (ii) the student; and notifying a developer of the tutoring agent of the identified possible improvement.

Abstract: Analyzing task associations may compute artifact associations for a plurality of artifacts based on historical task submission logs. From the computed artifact associations, shared artifacts that have artifact association with targeted artifacts that are associated with planned tasks, may be determined. Task association between the planned tasks may be identified based on the shared artifacts. Task associations may be analyzed to determine one or more of task assignment and task schedule for the planned tasks.

Abstract: Methods and arrangements for performing a tree-based asset search. An asset model is assimilated. The asset model is transformed into an asset tree search model. An asset search query is accepted and the asset search query is parsed. An asset search is conducted via applying the parsed asset search query to the asset tree model.

Abstract: Analyzing task associations may compute artifact associations for a plurality of artifacts based on historical task submission logs. From the computed artifact associations, shared artifacts that have artifact association with targeted artifacts that are associated with planned tasks, may be determined. Task association between the planned tasks may be identified based on the shared artifacts. Task associations may be analyzed to determine one or more of task assignment and task schedule for the planned tasks.

Abstract: Accommodating schedule variance in work, in one aspect, may comprise tracking information associated with work packets; identifying one or more work packets deviating from a planned schedule based on the tracking; identifying one or more features associated with the identified one or more work packets; computing metrics associated with the one or more features; prioritizing the identified one or more work packets based on the computed metrics using a predictive model, the predictive model calibrated at least based on historical data; and recommending one or more actions to take associated with the one or more prioritized work packets.

Abstract: Accommodating schedule variance in work, in one aspect, may comprise tracking information associated with work packets; identifying one or more work packets deviating from a planned schedule based on the tracking; identifying one or more features associated with the identified one or more work packets; computing metrics associated with the one or more features; prioritizing the identified one or more work packets based on the computed metrics using a predictive model, the predictive model calibrated at least based on historical data; and recommending one or more actions to take associated with the one or more prioritized work packets.

Abstract: A powerful set of features can assist software developers in searching source code. Source code can be queried to find occurrences of source code elements mapped to domain concepts appearing in an ontology. Queries can specify a search for occurrences of particular operations performed on domain concepts within the source code. Query expansion can be used to locate related domain concepts as indicated in the ontology. Query keyword weighting can be used to emphasize one domain concept over another. Tools can be provided to create the ontology and associate the ontology with the elements appearing in the source code. Results can include user interface features assisting in navigation to a location within the source code associated with the query results.

Abstract: The present invention describes a method and a computer program product for finding an impact of a change on a computer generated code. The method includes the generation of a dependency graph for the computer generated code. The dependency graph includes a plurality of nodes that are connected logically by one or more edges. The method enables an expert to assign nodal weights to the nodes, based on their importance. The method also enables the expert to assign edge weights and distribution ratios, based on the criticality of the relationship between the nodes and their distance from the root node. Thereafter, the impact of the change is found by implementing a spreading activation technique on the dependency graph. The output gathered after the activation represents the impact of the change on the computer generated code.

Abstract: Methods and arrangements for dynamic trendline generation. Input data are assimilated at a capture timepoint from an estimation tool. The input data are filtered, and an estimation-centric map is created from the filtered data ; this creating of a map includes generating an estimation data tree. A trendline is dynamically generated responsive to a request, wherein this generating of a trendline includes applying the estimation data tree to the input data. The dynamically generated trendline is supplied to the estimation tool.

Abstract: Methods and systems for automatically generating a service oriented architecture (SOA) design. A set of business process maps for the domain under consideration is defined and a design quality goal (function) that should be met (optimized) is specified. The design goal/function involves SOA metrics like coupling, cohesion, granularity, etc., which the system under consideration is pre-programmed to compute on any SOA design. The system takes as input the set of business process maps and the design quality goal/functions. It first generates semantic business process maps by identifying key concepts that occur in the task and business item descriptions. Next, it efficiently searches the service design space by starting with a seed design and employing a sequence of moves to iteratively optimize it. It outputs a set of possible SOA designs that meet the specified quality goals or optimizes the specified function, from where a user may select the final design.