Abstract: A technique for semantic search and retrieval that is event-based, wherein is event is composed of a sequence of observations that are user speech or physical actions. Using a first set of conversations, a machine learning model is trained against groupings of utterances therein to generate a speech act classifier. Observation sequences therein are organized into groupings of events and configured for subsequent event recognition. A set of second (unannotated) conversations are then received. The set of second conversations is evaluated using the speech act classifier and information retrieved from the event recognition to generate event-level metadata that comprises, for each utterance or physical action within an event, one or more associated tags. In response to a query, a search is performed against the metadata. Because the metadata is derived from event recognition, the search is performed against events learned from the set of first conversations.

Abstract: A method of fabricating a graphene-based solid-state device, the method including: disposing a graphene layer on a substrate; depositing a sacrificial layer on the graphene layer, the sacrificial layer being made of a non-polymeric dielectric material; patterning the graphene layer by defining at least one channel region, wherein the patterning is done by applying a lithographic process followed by an etching process using a resist layer, thus obtaining a patterned graphene layer protected against contamination from the resist layer by the sacrificial layer; patterning on the graphene layer a geometry of at least one metallic contact to be deposited; depositing the metallic contact on the graphene layer. A graphene-based solid-state device includes: a substrate; a graphene layer disposed thereon and defining at least one graphene channel; the graphene layer protected by a sacrificial layer made of a non-polymeric dielectric material; at least one metallic contact in contact with the graphene channel.

Abstract: A method of fabricating a graphene-based solid-state device includes: disposing a graphene layer on a substrate; depositing a first layer of a sacrificial material on the graphene layer, the first layer made of a first oxide dielectric material: patterning the graphene layer by defining at least one channel region, applying a lithographic process followed by etching; depositing a second layer of a sacrificial material on the stacked structure, the second layer made of a second oxide dielectric material different from the first. The selectivity to at least one etchant of the first oxide dielectric material is different from the selectivity to at least one etchant of the second. Also included is defining contact areas on which metal is deposited; etching away portions of the sacrificial material layers, defining hollow portions through which graphene areas are exposed; and in a single deposition stage, applying metal depositions on defined contact areas.

Abstract: A method of metallizing a solid-state device including a substrate and at least one graphene channel on the substrate includes applying a first metal deposition stage for depositing at least one first metallic structure, the metallic structure partially occupying the graphene layer and the substrate, wherein a metal deposition technique which does not use plasma is used to deposit the metallic structure; applying a second metal deposition stage for depositing at least one second metallic structure on a region of the first metallic structure not deposited on the graphene channel. A solid-state device includes at least two metallic structures, one partially occupying the graphene layer and the substrate and one partially deposited on a region of the at least one first metallic structure which is not deposited on the graphene channel. A method of packaging in a socket a die having at least one graphene-based solid-state device is related.

Abstract: A method of fabricating a graphene-based solid-state device include: disposing a layer of an electric conductive material a substrate; depositing a first layer on the electric conductive material, the first layer being made of a first oxide dielectric material; patterning the first layer to expose a portion of the layer of electric conductive material; disposing a graphene layer on the first layer; patterning the graphene layer to define a channel region; applying a lithographic process to define two contact areas in the graphene layer; depositing one metallic contact on each of the defined contact areas; depositing a second layer of an insulating material on the stacked structure, the second layer made of a second oxide dielectric material different from the first of which the first layer is made; and the selectivity to at least one etchant of the first oxide dielectric material is different from the second.

Abstract: A technique for semantic search and retrieval that is event-based, wherein is event is composed of a sequence of observations that are user speech or physical actions. Using a first set of conversations, a machine learning model is trained against groupings of utterances therein to generate a speech act classifier. Observation sequences therein are organized into groupings of events and configured for subsequent event recognition. A set of second (unannotated) conversations are then received. The set of second conversations is evaluated using the speech act classifier and information retrieved from the event recognition to generate event-level metadata that comprises, for each utterance or physical action within an event, one or more associated tags. In response to a query, a search is performed against the metadata. Because the metadata is derived from event recognition, the search is performed against events learned from the set of first conversations.

Abstract: A method and computing platform to imitate human conversational response as a context transitions across multiple channels (e.g., chat, messaging, email, voice, third party communication, etc.) where inputs to the system are categorized into identified speech acts and physical acts, and a conversational bot is associated to the channels. In this approach, a data model associated with a multi-turn conversation is provided. The data model comprises an observation history, wherein an observation in the observation history includes an identification of a channel in which the observation originates. As turns are added to the multi-turn conversation, a conversational context across multiple channels is persisted using the data model. Using this approach, an AI-supported conversation started in one channel can move to another conversation channel while maintaining the context of the conversation intact and coherent.

Abstract: A method and computing platform to imitate human conversational response as a context transitions across multiple channels (e.g., chat, messaging, email, voice, third party communication, etc.) where inputs to the system are categorized into identified speech acts and physical acts, and a conversational bot is associated to the channels. In this approach, a data model associated with a multi-turn conversation is provided. The data model comprises an observation history, wherein an observation in the observation history includes an identification of a channel in which the observation originates. As turns are added to the multi-turn conversation, a conversational context across multiple channels is persisted using the data model. Using this approach, an AI-supported conversation started in one channel can move to another conversation channel while maintaining the context of the conversation intact and coherent.

Abstract: A technique for semantic search and retrieval that is event-based, wherein is event is composed of a sequence of observations that are user speech or physical actions. Using a first set of conversations, a machine learning model is trained against groupings of utterances therein to generate a speech act classifier. Observation sequences therein are organized into groupings of events and configured for subsequent event recognition. A set of second (unannotated) conversations are then received. The set of second conversations is evaluated using the speech act classifier and information retrieved from the event recognition to generate event-level metadata that comprises, for each utterance or physical action within an event, one or more associated tags. In response to a query, a search is performed against the metadata. Because the metadata is derived from event recognition, the search is performed against events learned from the set of first conversations.

Abstract: A technique for semantic search and retrieval that is event-based, wherein is event is composed of a sequence of observations that are user speech or physical actions. Using a first set of conversations, a machine learning model is trained against groupings of utterances therein to generate a speech act classifier. Observation sequences therein are organized into groupings of events and configured for subsequent event recognition. A set of second (unannotated) conversations are then received. The set of second conversations is evaluated using the speech act classifier and information retrieved from the event recognition to generate event-level metadata that comprises, for each utterance or physical action within an event, one or more associated tags. In response to a query, a search is performed against the metadata. Because the metadata is derived from event recognition, the search is performed against events learned from the set of first conversations.

Abstract: A displayed document comprises an annotation widget, the widget associated with an annotation document and a corresponding annotation key in an annotation store. The annotation document associated with a workflow action program. A user with a predetermined privilege selects a widget and is presented with the annotation document. The user performs an annotation task modifying the annotation document and submits the annotation document to the annotation store, the submission triggering the workflow action program to progress the workflow to another step.

Abstract: Generating a unique URL key for a web document according to an obtained key generating policy. The URL of a web page is parsed according to the policy in order to generate the URL key. Preferably, the key generating policy is obtained from a well known source associated with the web page.

Abstract: Digital Fingerprints are generated for data objects in a system where separate annotation files are created for data objects. This permits cross heterogeneous system relationship of a data object with associated annotations. The digital fingerprint is saved in an annotation store along with a first relationship between the digital fingerprint and the location of annotations as well as a second relationship between the digital fingerprint and location of copies of the data object. The digital fingerprint can be generated by any system that has a copy of the data object. Annotations or data objects can be found by searching for the digital fingerprint and its relationships.

Abstract: In response to a user request, an annotation form is presented to a user for annotating a target document or image in a database. The annotation form is associated with the target document. A runtime program is associated with the annotation form. The annotation form selected by the user provides prompts for annotation input. The annotation input is saved in an annotation database and the runtime program performs an action on behalf of the user. Preferably the runtime program action comprises analysis of elements of the target document.

Abstract: A displayed document comprises an annotation widget, the widget associated with an annotation document and a corresponding annotation key in an annotation store. The annotation document associated with a workflow action program. A user with a predetermined privilege selects a widget and is presented with the annotation document. The user performs an annotation task modifying the annotation document and submits the annotation document to the annotation store, the submission triggering the workflow action program to progress the workflow to another step.

Abstract: A displayed document comprises an annotation widget, the widget associated with an annotation document and a corresponding annotation key in an annotation store. The annotation document associated with a workflow action program. A user with a predetermined privilege selects a widget and is presented with the annotation document. The user performs an annotation task modifying the annotation document and submits the annotation document to the annotation store, the submission triggering the workflow action program to progress the workflow to another step.

Abstract: Digital Fingerprints are generated for data objects in a system where separate annotation files are created for data objects. This permits cross heterogeneous system relationship of a data object with associated annotations. The digital fingerprint is saved in an annotation store along with a first relationship between the digital fingerprint and the location of annotations as well as a second relationship between the digital fingerprint and location of copies of the data object. The digital fingerprint can be generated by any system that has a copy of the data object. Annotations or data objects can be found by searching for the digital fingerprint and its relationships.

Abstract: Digital Fingerprints are generated for data objects in a system where separate annotation files are created for data objects. This permits cross heterogeneous system relationship of a data object with associated annotations. The digital fingerprint is saved in an annotation store along with a first relationship between the digital fingerprint and the location of annotations as well as a second relationship between the digital fingerprint and location of copies of the data object. The digital fingerprint can be generated by any system that has a copy of the data object. Annotations or data objects can be found by searching for the digital fingerprint and its relationships.

Abstract: A method, system, and computer program product for managing at least one work flow in a distributed system are disclosed. The method includes receiving a notification from a central storage system identifying a participant's information processing system as at least one participant in at least one workflow. The at least one workflow is analyzed to determine if at least one step of the workflow is currently executable at the participant's information processing system. The step is executed in response to determining that the step is currently executable at the participant's information processing system. Workflow update information is transmitted to the central storage system for association with the workflow.

Abstract: A method, apparatus, and computer program product for graphically presenting compatible workflow steps to a user through a graphical user interface are disclosed. The method includes graphically displaying a plurality of steps in a workflow. A user's selection for at least one step in a plurality of steps to denote at least one selected step and at least one non-selected step is received. The method further includes determining if a set of one or more output parameters of the selected step is acceptable as an input of the non-selected step. An appearance of the non-selected step is graphically changed if the output parameters are acceptable. The method also includes determining if a set of one or more output parameters of the non-selected step is acceptable as an input of the selected step. An appearance of the non-selected step is graphically changed if the output parameters are acceptable.