Abstract: The methods, systems, and computer program products described herein provide optimized provisioning of replacement parts for service calls through the use of machine learning. In some aspects, historical hardware maintenance tickets may be processed to generate symptoms vectors identifying sets of symptoms associated with the hardware maintenance tickets. The symptoms vectors and corresponding parts usage records of the historical hardware maintenance tickets may be used train a decision model to predict a probability that a particular part will be used to fulfill the new hardware maintenance ticket. The predicted probability may be used by the system when generating a parts provisioning plan for the new hardware maintenance ticket.

Abstract: Procedure dialogs are improved through knowledge mining within a reinforcement learning framework. Taking an existing procedure dialog as input, a machine learning model is generated. User interactions with the machine learning model are monitored and used to update the machine learning model. The updates to the machine learning model are applied to the existing procedure dialog for review and revision by subject matter experts.

Abstract: Integration of personalized traffic information is provided. Contextual information for a client device is collected. Based, at least in part, on the contextual information, traffic information is received. The traffic information is received from one or more broadcast stations to which the traffic information has been mapped based, at least in part, on classification of the traffic information. A communications device is caused to communicate the traffic information.

Abstract: A computer-implemented method, apparatus, computer program product for intervention in a conversation between a virtual agent and a user is disclosed. In the computer-implemented method, a risk level of the conversation is evaluated. In response to the risk level being higher than a threshold, it is indicated that the conversation needs an intervention by a human agent. In response to the human agent intervening in the conversation, the conversation is handed over from the virtual agent to the human agent. Then, a simulative conversation is generated according to the conversation between the user and the human agent. In response to a determination by the human agent based on the simulative conversation, the conversation is handed over from the human agent to the virtual agent.

Abstract: An authentication method, system, and computer program product for decoding brain activity into a brain pattern sequence, applying brain dialogue to interact with a user while the user is entering the brain pattern sequence, and matching the brain pattern sequence with a predetermined password to allow access to a system.

Abstract: A system and method to build feature evolution models for existing applications (“apps”) in the market based on open app data repositories). A feature evolution model of an app depicts the app name, its historical versions (historical version labels, release timestamps of each version), rating values of each version, and structured features (e.g., umbrella features and low-level features) each version introduces, improves or deletes. There is further extracted from the app description and release logs the app name, historical version labels, release timestamps, use the rating info of the app to extract and assign rating values for each version of the app, and apply NLP techniques and source code analysis techniques to extract “structured features” of the app through analyzing the app description, the release logs, and corresponding source code revisions of the app. Upon the built feature evolution models, various feature insights may be easily extracted and generated.

Abstract: A method, system, and computer program product, include obtaining window status decision information based on vehicle interior background noise and determining switch status of the vehicle window(s) based on the obtained window status decision information.

Abstract: Procedure dialogs are improved through knowledge mining within a reinforcement learning framework. Taking an existing procedure dialog as input, a machine learning model is generated. User interactions with the machine learning model are monitored and used to update the machine learning model. The updates to the machine learning model are applied to the existing procedure dialog for review and revision by subject matter experts.

Abstract: Aspects of the present invention generally relate to a system, method, and computer program product for transforming APIs from legacy system-of-record to new system-of-engagement using complete API process logic. API candidates are generated from execution traces for transforming the API from the legacy system to a new system of record. An embodiment of the invention is first, add implementers into a legacy systems and collect the instrumenter output as execution traces. Second, rank the execution traces and find the most frequent execution traces. Third, consolidate the execution traces by merging common trace segments. The codes covered by the consolidated execution traces will be considered as candidate API components. Fourth, expand the merged execution traces and find related execution traces and check the connectivity between different nodes.

Abstract: A system and method to build feature evolution models for existing applications (“apps”) in the market based on open app data repositories). A feature evolution model of an app depicts the app name, its historical versions (historical version labels, release timestamps of each version), rating values of each version, and structured features (e.g., umbrella features and low-level features) each version introduces, improves or deletes. There is further extracted from the app description and release logs the app name, historical version labels, release timestamps, use the rating info of the app to extract and assign rating values for each version of the app, and apply NLP techniques and source code analysis techniques to extract “structured features” of the app through analyzing the app description, the release logs, and corresponding source code revisions of the app. Upon the built feature evolution models, various feature insights may be easily extracted and generated.

Abstract: A mechanism is provided for optimizing web page pre-loading. Source codes of a web page file are analyzed to generate a module dependency graph that describes load dependency relationships between respective modules in the web page file. A load flow graph that indicates load flows of respective modules in the web page file is generated based on the module dependency graph. The load flow graph and the module dependency graph are analyzed to determine pre-load sequences of respective modules in the web page file. A pre-load profile is then generated based on the determined pre-load sequences.

Abstract: A method, system, and computer program product, include obtaining window status decision information based on vehicle interior background noise and determining switch status of the vehicle window(s) based on the obtained window status decision information.

Abstract: Method and systems for tagging code include organizing code units in a hierarchical structure that associates the code units with respective description units. The hierarchical structure is filtered to remove un-meaningful description units from the hierarchical structure. Feature tags are generated based on a weighted sum of term frequencies in the description units of the hierarchical structure to associate features with respective code units.

Abstract: A method, system, and computer program product, include obtaining window status decision information based on vehicle interior background noise and determining switch status of the vehicle window(s) based on the obtained window status decision information.

Abstract: A vehicle occupant comfort system, comprises a processor that stores computer executable components stored in memory. A plurality of sensors sense ambient conditions associated with exterior and interior conditions of a vehicle. A context component infers or determines context of an occupant of the vehicle. A comfort model component implicitly and explicitly trained on occupant comfort related data analyzes information from the plurality of sensors and context component. A comfort controller adjusts environmental conditions of a passenger compartment of the vehicle based at least in part on output of the comfort model component.

Abstract: Techniques for response generation are provided. One computer-implemented method comprises: receiving, by an electronic device operatively coupled to a processing unit, a query from a user. The computer-implemented method also comprises generating, by the electronic device, a dialog candidates set based on user context, wherein the user context comprises the query. The computer-implemented method can also comprise generating, by the electronic device, a response based on the dialog candidates set.

Abstract: Method and systems for tagging code include building a code structure based description unit association model (CDAM) that associates code units with respective description units and organizes the code units in a hierarchical structure. The CDAM is filtered to remove un-meaningful description units through cross-validation between code units. Feature tags are generated based on a weighted sum of term frequencies in the description units of the CDAM to associate features with respective code units.

Abstract: A method for predicting anomalies in a computer application includes during runtime of the computer application, detecting traffic metrics and incident tickets associated with the computer application, the incident ticket indicating an incident might occur in the computer application; calculating a threshold based on absolute values of second order differences associated with the traffic metrics, wherein the threshold is such that when the absolute value of the second order difference associated with the traffic metrics exceeds the threshold, a recall rate Rrecall that the computer application is recalled is maximized; obtaining predicted metrics of the computer application in a next time period based on the traffic metrics; and in response to an absolute value of a second order difference associated with the predicted metrics exceeding the threshold, predicting potential anomalies of the computer application in the next time period.

Abstract: A system and method to build feature evolution models for existing applications (“apps”) in the market based on open app data repositories). A feature evolution model of an app depicts the app name, its historical versions (historical version labels, release timestamps of each version), rating values of each version, and structured features (e.g., umbrella features and low-level features) each version introduces, improves or deletes. There is further extracted from the app description and release logs the app name, historical version labels, release timestamps, use the rating info of the app to extract and assign rating values for each version of the app, and apply NLP techniques and source code analysis techniques to extract “structured features” of the app through analyzing the app description, the release logs, and corresponding source code revisions of the app. Upon the built feature evolution models, various feature insights may be easily extracted and generated.

Abstract: A system and method to build feature evolution models for existing applications (“apps”) in the market based on open app data repositories). A feature evolution model of an app depicts the app name, its historical versions (historical version labels, release timestamps of each version), rating values of each version, and structured features (e.g., umbrella features and low-level features) each version introduces, improves or deletes. There is further extracted from the app description and release logs the app name, historical version labels, release timestamps, use the rating info of the app to extract and assign rating values for each version of the app, and apply NLP techniques and source code analysis techniques to extract “structured features” of the app through analyzing the app description, the release logs, and corresponding source code revisions of the app. Upon the built feature evolution models, various feature insights may be easily extracted and generated.