Abstract: A system and method for creating a machine learning (ML) classifier for a database uses a weakly-supervised training data set created automatically from database items on the basis of a human-created keyword set. The automatically created training data set is used to construct one or more deep learning classifier checkpoints, which can then be compared with one another and with a classifier based on the original keyword set in order to select a classifier for use by other users viewing the database.

Abstract: One example provides a computing system comprising a depth sensor comprising a plurality of pixels, and a storage machine holding instructions executable by a logic machine to, for each pixel, make K phase measurements to form a set of noisy phase measurements, determine a location at which a projection line that passes through the set of noisy phase measurements in a K-dimensional phase space passes through a lower dimensional plane, the projection line being parallel to a noise free phase evolution line, compare the location to a plurality of independent terms of a predetermined matrix of points in the lower dimensional plane, locate a corresponding set of noiseless phase orders by using a selected set of independent terms to reference a look-up table, determine a distance value for the pixel based upon the corresponding set of noiseless phase orders, and output the distance value for the pixel.

Abstract: A training process produces a machine-learned model that, once trained, can be applied to process different types of data items. The training process accomplishes this result by combining data items in a training set with type-specific parameter information, to produce supplemented data items. The training process then trains a model based on the supplemented data items. Training involves adjusting model weights together with the type-specific parameter information. In an inference stage of processing, the technology combines a new data item with an appropriate type of trained parameter information, and then maps the resultant supplemented data item to an output data item. The technology is particularly effective in adapting an initial model to a new subject matter domain in those situations in which a robust set of data items that pertain to the subject matter domain and which have a desired type is lacking.

Abstract: A foldable computer mouse is provided that includes a deformable body formable into an expanded configuration usable for receiving inputs for controlling a computing device and a folded configuration. The expanded configuration being an arcuate configuration having a first curvature and the folded configuration having a second curvature greater than the first curvature. The mouse also includes a hollow shell disposed on a top surface of the deformable body. The shell expands based at least on the deformable body being formed into the expanded configuration, and the shell collapses against the top surface of the deformable body based at least on the deformable body being folded into the folded configuration. The shell comprises a flexible material that retains a shape of the shell in the expanded configuration.

Abstract: According to examples, an apparatus may include a processor that may determine that an application was accessed through a portal. Based on a determination that the application was accessed through the portal, the processor may determine whether a first credential type or a second credential type was supplied to access the application, in which the first credential type may include a set of personal credentials of a user and the second credential type may include a set of single sign-on credentials that the user may use to access multiple applications. The processor may also output a trace that may indicate an identification of the application that was accessed and the type of the credential supplied to access the application, in which a backed entity may analyze the data included in the trace.

Abstract: Examples are disclosed relating to providing haptic output to a stylus. In one example, rotational position data indicating a rotational position of the stylus about a longitudinal axis of the body of the stylus is received. Travel direction data indicating a direction of travel of a tip of the stylus relative to a touch-sensitive screen of a computing device is also received. Using at least the rotational position data and the travel direction data, one or more characteristics of a drive signal are determined. The drive signal is then transmitted to a haptic feedback mechanism within the body of the stylus to generate haptic output at the body.

Abstract: Methods and systems are disclosed for sharing a content item from a secondary computing device to a primary computing device based on a tilt position of the secondary computing device. A sensor on the secondary computing device determines that the secondary computing device has a first tilt position that is associated with a non-sharing mode. The sensor senses that the secondary computing device is tilting from the first tilt position towards a second tilt position that is associated with a full sharing mode. In response, the content item is begun to be shared with the primary computing device. An amount of the content item that is shared with the primary computing device is continued to be increased as long as the sensor continues to sense that the secondary computing device is tilting from the first tilt position towards the second tilt position. The sensor senses that the secondary computing device has reached the second tilt position and shares the entire content item with the primary computing device.

Abstract: According to examples, an apparatus may include a processor and a memory on which is stored machine-readable instructions that when executed by the processor, may cause the processor to determine that an entity was granted an anomalous role assignment to a managed environment. The processor may also, based on the determination that the role assignment of the entity is anomalous, identify at least one indicator associated with the role assignment, determine an indicator value corresponding to the identified at least one indicator, and determine whether the indicator value exceeds a predefined threshold value. The processor may, based on a determination that the indicator value exceeds the predefined threshold value, output an alert indicating that the role assignment is suspicious.

Abstract: A laser device includes a laser and a controller. The laser has an optical cavity that includes an active gain section and a phase shifter. The controller is configured to excite the active gain section to lase light out of the optical cavity. The controller is further configured to, while the light is being lased out of the optical cavity, modulate a refractive index of the phase shifter to shift an optical phase of lasing modes of the lased light to thereby reduce coherence of the lased light.

Abstract: Computing devices and methods for determining opening and closing of touch sensitive interfaces are disclosed. In one example, a computing device comprises a touch screen display on a first substrate that is rotatably coupled to a second substrate that includes a trackpad. A trackpad identification signal transmitted by the trackpad is received at the touch screen display, and a touch screen identification signal transmitted by the touch screen is received at the trackpad. If the trackpad identification signal matches a trackpad identification key and the touch screen identification signal matches a touch screen identification key, then an energy level of one or both signals is compared to an energy level threshold. Based at least in part on the comparison of the energy level to the threshold, a power state transition is initiated.

Abstract: A system to detect an abnormally permissive role definition, which can include an abnormally permissive custom role definition, and take action is described. The system receives a role definition for a security principal over a scope of resources in which the role definition includes a built-in role and a custom role. Permissions of the role definition and a creation event of the role definition are analyzed. A security score based on the role definition and creation event for the scope of resources is determined. An action is taken based on the security score and the creation event analysis.

Abstract: A touch-sensitive display device includes one or more touch-sensitive displays each including a plurality of touch-sensitive electrodes. A free touch-sensitive electrode is identified on the one or more touch-sensitive displays that is (1) at least temporarily unaffected by proximity of one or more input objects to the one or more touch-sensitive displays, and (2) affected by an electrical noise caused by display of image content on the one or more touch-sensitive displays. The electrical noise affecting the free touch-sensitive electrode is measured. Based at least in part on the measured electrical noise affecting the free touch-sensitive electrode, and using a trained neural network, an amount of electrical noise caused by the display of image content that is affecting an occupied touch-sensitive electrode is estimated, the occupied touch-sensitive electrode being affected by proximity of the one or more input objects to the one or more touch-sensitive displays.

Abstract: A dynamic peak power management system may prevent brownouts while improving performance and user experience compared to conventional techniques. A current threshold may be set below the maximum current capability (Imax) of a battery. If the current drawn from the battery exceeds the current threshold repeatedly, then system components may be throttled to decrease their peak power usage. If the current drawn from the battery stays below the current threshold for some time, then system components may be unthrottled to improve performance. This dynamic adaptable technique for managing peak power does not unnecessarily sacrifice performance by preemptively throttling system components to avoid the rare worst-case scenario where power spikes of system components perfectly align in time.

Abstract: A computer-implemented method for implementing a map calendar graphical user interface (GUI) including content-variable view levels is described herein. The method includes executing a map calendar application to provide a map calendar GUI surfaced on a display. The map calendar GUI includes interactive visualizations of a map calendar that are organized into content-variable view levels that relate to specific units of time.

Abstract: A system and method for automatically scaling consumer servers in a data processing system. To build an automatic scaling system, the present disclosure allows consumers to obtain additional information, e.g., the number of events that await to be read from an aggregator when receiving an event from the aggregator. This additionally obtained number provides a direct gauge for the data processing system to determine when the consumers are over-provisioned, i.e., when the number of events left to be read is close to zero, as well as when the consumers are under-provisioned, e.g., when the number of events left to be read continues to increase. As a result, the consumers can be automatically scaled to handle the dynamic data processing demand while providing optimal resource allocation.

Abstract: Methods and systems for generating multi-operator data transformation pipelines. An example method includes accessing raw data for transformation; receiving a selection of a target table or target visualization, wherein the target table or target visualization is for data other than the raw data; extracting table properties and target constraints; and based on the extracted table properties and target constraints, synthesizing one or more multi-operator data transformation pipelines for transforming the raw data to a generated table or generated visualization.

Abstract: A system and method for creating a machine learning (ML) classifier for a database uses a weakly-supervised training data set created automatically from database items on the basis of a human-created keyword set. The automatically created training data set is used to construct one or more deep learning classifier checkpoints, which can then be compared with one another and with a classifier based on the original keyword set in order to select a classifier for use by other users viewing the database.

Abstract: According to examples, an apparatus may include a processor and a memory on which is stored instructions that may cause the processor to determine that information has been inputted into an object, the object including a child object nested within the object. The instructions may also cause the processor to generate a child-object hash value based on contents of the child object, generate a hash value for the object based on contents of the object, the contents of the object including the child-object hash value, and determine whether the generated hash value differs from a stored hash value, the stored hash value corresponding to contents of the object without the inputted information. The instructions may also cause the processor to, based on a determination that the generated hash value differs from the stored hash value, save the object including the child object nested within the object.

Abstract: A method for providing automatic restatement of a data environment includes receiving a request for restatement of a first dataset, accessing dependency data that describes a set of dependencies between the first dataset and one or more other datasets in the computer data environment, examining the dependency data to automatically identify one or more second dataset from among the one or more other datasets, the second datasets being datasets from which the first dataset depends, examining a first timing data related to the restatement of the first dataset and a second timing data related to the restatement of the identified second datasets to identify one or more third datasets from among the second datasets that require restatement as a result of the restatement of the first dataset, generating a restatement plan based on the identified third datasets, and initiating automatic execution of the restatement plan.

Abstract: A system and method for online user feedback management are provided. The method includes receiving online user feedbacks for a product from a plurality of users. A plurality of topics for the product are identified from the online user feedbacks. For each topic, the received online user feedbacks are categorized into a plurality of groups based on a rating score provided in each online user feedback for the product and semantic analysis of each online user feedback for the product. A net promoter score (NPS) uplift for each topic is calculated, where the NPS uplift measures an improvement in a NPS for the product if issues related to the topic are resolved. A priority topic is identified based on the NPS uplift for each of the topics. The priority topic is then prioritized in resolving issues related to the topics included in the online user feedbacks.

Abstract: The present disclosure relate to highlighting audience members with reactions to a presenter of an online meeting. Unlike physical, fact-to-face meeting that enables spontaneous interactions among the presenter and the audiences that are collocated with the presenter, presenting materials during an online meeting raises an issue of the present not being able to see real-time reactions or feedback by the audience members. The present disclosure addresses the issue by dynamically determining one or more audience members who indicate reactions during the online meeting or presentation and displaying faces of the one or more audience members under spotlight to the presenter. The presenter sees faces of the audience members with reactions during the online presentation and responds to the audience members and keep the audience engaged. The spotlight audience server analyzes video frames and determines types of reactions of the audience members.

Abstract: Examples are disclosed herein that relate to creating undercuts and other complex geometries in an injection molded article. One example provides a method of forming an injection molded article comprising an undercut structure, the method comprising inserting an insert piece into a mold for forming the molded article, the insert piece comprising sacrificial material, injection-molding the molded article, thereby incorporating the insert piece into the injection molded article, and machining the insert piece to form the undercut structure from the sacrificial material.

Abstract: A method for providing generating data extensions for a first data profile includes receiving a request for historical data, the data being associated with a first data profile, the first data profile being generated from data supplied by a data source system, creating a data model for obtaining the historical data based on data from at least one of the first data profile, a second data profile, or the data source system, executing the data model to obtain the historical data, and storing the historical data as a data extension of the first data profile. The first data profile and the data extension of the first data profile may be generated and stored separately, and the data extension of the first data profile is generated on the fly at runtime.

Abstract: A method for a computing device. The method includes periodically transmitting a beaconing data packet via a first wireless communication protocol. The first wireless communication protocol supports low bandwidth, one-directional communication. The computing device receives a request to pair from a peripheral computing device via a second wireless communication protocol. The second wireless communication protocol supports higher bandwidth, two-directional communication. The request to pair includes information that was included in the beaconing data packet. Responsive to receiving the request to pair, the computing device initiates pairing with the peripheral computing device via the second wireless communication protocol.

Abstract: A computing device running a tool for visually exploring datasets receives a request for a dataset comprising a plurality of subspaces of a multi-dimensional data structure and in response to receiving a user selection of a data point associated with the dataset, presents a visualization of the calculations associated with the selected data point. The tool allows the user to select one of the calculations, thus causing presentation of a visualization of the selected calculation. A semantic graphical feature is incorporated into the visualization of the selected calculation depending upon a type of the calculation.

Abstract: A computer-implemented technique uses a machine-trained model to convert input information provided by a user into a natural language expression and a semantic vector. The semantic vector provides a distributed representation of the natural language expression. The technique then uses the generated semantic vector to find a target data record having a target semantic vector that is within a prescribed distance to the generated semantic vector. The technique then outputs at least a target natural language expression that is associated with the target data record. The target data record is produced in an offline process that converts one or more attribute values in an existing structured data record into the target semantic vector and the target natural language expression. Also described herein is a technique for training the machine-trained model and for updating existing collections of data records using the machine-trained model.

Abstract: A topic-processing system processes topics in a set of documents in a two-stage manner. In the first stage, the system recognizes candidate topics in the set of documents using a machine-trained named-entity recognition (NER) model, to produce original NER information. In a second stage, the system aggregates the original NER information over the set of documents, to produce aggregated information. The system then ranks the candidate topics in the set of candidate topics based on the aggregated information using a machine-trained classification model, to produce a set of ranked topics. The system then selects a set of final topics from the set of ranked topics, e.g., by excluding ranked topics having scores below a prescribed threshold value. A production system presents supplemental information regarding selected final topics, where those final topics are identified by the topic-processing system.

Abstract: A circuit configured to transmit frequency multiplexed signals from a superconducting domain to a higher temperature domain. The circuit comprising a multiplexed signal output and a plurality of superconducting oscillator circuits each configured to output a different carrier frequency, each superconducting oscillator circuit comprising an oscillator output connected to the multiplexed signal output. Each superconducting oscillator circuit comprising a converter stage configured to convert an input of a superconducting logic signal to a Single Flux Quantum (SFQ) bit value, a splitter stage electrically connected to an output of the converter stage, the splitter stage configured to change between a first current state and a second current state based at least in part on the SFQ bit value, and an oscillator stage magnetically coupled to an output of the splitter stage and electrically coupled to the oscillator output.

Abstract: Email has become commonplace as a means to communicate among parties non-synchronously by exchanging email messages. Some communications, however, may be more effective in real-time or in near real-time. For instance, service providers may offer a chat feature to accommodate real-time or near real-time help or service communications with customers. However, in this case, the user is likely required to enter credentials and/or context information to transition into a different communication type with the service provider. The present systems and methods provide a continuous conversation experience for the user whereby a message recipient of one message type (e.g., email, text, etc.) may continue a conversation in a different message type (e.g., a chat session, instant message, etc.) without manually providing message context for communicating via the second message type.

Abstract: A method of reducing power consumption of a first wireless communication device is described. A charge level of a battery associated with the first wireless communication device is monitored. A wireless communication session between the first wireless communication device and a second wireless communication device is maintained. Based at least in part on the charge level of the battery being within a low battery threshold range, a wireless signal strength associated with the wireless communication session is monitored. Based at least in part on the wireless signal strength reaching a power saving threshold that is above a minimum connection threshold for maintaining the wireless communication session, a power saving action associated with a wireless interface that supports the wireless communication session is performed.

Abstract: According to examples, an apparatus may include a processor and a memory on which are stored machine-readable instructions that when executed by the processor, may cause the processor to receive information regarding a property of a data resource and to calculate a score associated with the data resource based on the received information regarding the property. The score may correspond to a probability that the data resource is to have a first type of accessibility. The processor may compare the calculated score to a predefined threshold value and, based on the comparison of the calculated score to the predefined threshold value, may output an indication that accessibility to the data resource should be set to the first type of accessibility or a second type of accessibility.

Abstract: The discussion relates to automatically providing information about what code sequences contribute to a length of time a program takes to execute. One example can collect context switch and ready thread event tracing data from a program over a period of interest and identify time blocks of program threads from the period of interest. The example can distinguish individual time blocks that contribute to execution time for the period of interest from other individual time blocks that do not contribute to the execution time.

Abstract: Systems and methods for contextual memory capture and recall are provided. The contextual memory capture and recall systems and methods help a user create, store, and recall memory information associated with an identified activity. The contextual memory capture and recall systems and methods are capable of identifying user activities where a memory inquiry may be desirable, creating a memory inquiry with a recommended memory action based on the activity, providing the memory inquiry to the user, and automatically linking an accepted memory action, along with any received memory information for the memory action, with the identified activity.

Abstract: A method of training a neural network for detecting target features in images is described. The neural network is trained using a first data set that includes labeled images, where at least some of the labeled images having subjects with labeled features, including: dividing each of the labeled images of the first data set into a respective plurality of tiles, and generating, for each of the plurality of tiles, a plurality of feature anchors that indicate target features within the corresponding tile. Target features that correspond to the plurality of feature anchors are detected in a second data set of unlabeled images. Images of the second data set having target features that were not detected are labeled. A third data set that includes the first data set and the labeled images of the second data set is generated. The neural network is trained using the third data set.

Abstract: Examples of the present disclosure describe systems and methods for securely cascading build pipelines to various software engines, hardware platforms, and/or devices. In aspects, a set of input parameters may be received by a software automation and/or development system. Based on the set of input parameters, the system may create a software build job to test software for various build pipelines. Upon execution of the software build job, the system may identify a set of software engine-specific or hardware platform-specific software build devices. The system may cause the set of software build devices to concurrently build respective software. One or more test programs of the software may be executed, and the corresponding test results may be provided to a results location. The system may securely access the test results in the result location(s) and present the test results for the software build job in a single user interface.

Abstract: Some storage systems are configured with VDL (valid data length) type controls that are implemented on a per cluster basis and, in some instances, on a sub-cluster basis, rather than simply a per file basis. In some instances, per-cluster VDL metadata for the storage clusters is stored and referenced at the edge data volume nodes of a distributed network for the storage system rather than, and/or without, storing or synchronizing the per-cluster VDL metadata at a master node that manages the corresponding storage clusters for the different data volume nodes. Sequence controls are also provided and managed by the master node and synchronized with the edge data volume nodes to further control access to data contained in the storage clusters.

Abstract: An intelligent agent may assume a ghost presence in a meeting on behalf of a non-participant, monitor data communications between participants of the meeting, and identify appropriate triggers or events relating to the non-participant. For example, the agent may detect questions directed at the non-participant or that the non-participant has the knowledge to answer. The non-participant may be auto-summoned with respect to the meeting when the events relating to the non-participant are identified. The auto-summoning may be performed by communicating information about identified event to the non-participant, for example in real-time. The communication about the identified event may be transmitted over any of a variety of communication channels or modalities, including instant message chat, SMS, phone call, establishing a video call, or pager.

Abstract: A productivity entity container data structure is created that includes information for a first productivity entity and a link to a second productivity entity. The information for the first productivity entity includes either a link to the first productivity entity or details for the first productivity entity stored directly in the productivity entity container data structure. The first productivity entity and the second productivity entity are of different productivity entity types (e.g., an assigned task associated with a scheduled calendar event). By using the productivity entity container data structure to link together multiple different productivity entities of different productivity entity types, computer-based productivity systems are configured to directly access and display information for related productivity entities of different productivity entity types in a unified-view interface.

Abstract: A wearable computing device, including a device body, an inertial measurement unit (IMU), and a processor. The processor may receive, from the IMU, a plurality of kinematic measurements collected within a time window. With one or more machine learning models, based at least in part on the kinematic measurements, the processor may compute a current velocity estimate for the wearable computing device at a current timestep and a prior velocity estimate for the wearable computing device at a prior timestep. The processor may compute a current pose estimate and a prior pose estimate based at least in part on the current velocity estimate and the prior velocity estimate, respectively. The processor may compute a composite pose estimate for the wearable computing device at the current timestep based on the current pose estimate and the prior pose estimate. The processor may output the composite pose estimate to a target program.

Abstract: The present disclosure relates to systems and methods for an interactive, intelligent hub built around the completion of a task. This hub brings together resources, information, suggested steps, and other automated assistance to facilitate the completion of the task. AI-based assistance may indicate which steps can be completed by automated processes, and dispatch those processes, or suggest resources to assist in the completion of other steps. The hub displays the current status of the task, and lives until the completion of the task, or abandonment by the user.

Abstract: A system to determine an intrusion risk and take action is described. The system collaboratively filters a combination based on a user access and a network item in a computer network to determine an associated recommendation score. The system determines connected components of a model of the computer network and separately collaboratively filters the connected components to determine the recommendation score as a measure of intrusion risk. An action is taken on the user access based on the intrusion risk.

Abstract: The description relates to cooling electronic components, such as computing devices. One example includes a rack defining a volume and multiple sealed chassis modules removably fluidly coupled to a two-phase condenser tank via a vapor coupler and a liquid coupler. Individual sealed chassis modules can contain one or more electronic components immersed in two-phase coolant that when heated by operation of the electronic components experiences a phase change from a liquid phase to a gas phase and travels to the two-phase condenser tank via the vapor coupler and is cooled in the two-phase condenser tank until experiencing a phase change back into the liquid phase. Individual sealed chassis modules can be decoupled from the two-phase condenser tank without releasing two-phase coolant and an entirety of the multiple sealed chassis modules and the condenser tank are contained in the volume of the rack.

Abstract: Examples of the present disclosure describe systems and methods for establishing a persistent connection between file attachments uploaded to cloud storage and associated electronic communications. An electronic communication including an attached file may be stored and presented in a communication application user interface that includes interactive functions, such as edit and save functions. Upon selection of an interactive function, a copy of the file may be uploaded to a cloud storage location and, in some examples, opened in a productivity application. A link to the file at the cloud storage location may be stored as a property of the file that is stored with the electronic communication. Resultantly, a connection may be maintained between the file at the cloud storage location and the electronic communication. Based on this connection, the copy of the file at the cloud storage location may be opened upon a subsequent selection of an interactive function.

Abstract: A computer-implemented method includes receiving, by a proxy device, a document from a service provider in response to a request to the service provider from a client device. The proxy device injects into the document event monitoring code for monitoring user actions on the client device. The proxy device sends the document with the event monitoring code to the client device. The event monitoring code intercepts a user request for a file upload event using a client-side application on the client device. The proxy device receives a client request including file information regarding the file upload event from the event monitoring code. The proxy device determines whether the file upload event should be allowed or blocked based on the received file information and stored policy data.

Abstract: Techniques of configuring settings for automatic update of calendar items in an electronic calendar systems are disclosed herein. In one embodiment, a server can receive and store update configuration records indicating that a change to a start date, start time, an end date, an end time, or location of a meeting would render a meeting update actionable. Upon receiving an incoming meeting update, the server can determine whether the incoming meeting update contains a change to one of a start date, start time, an end date, an end time or location. If true, the server can automatically modify, without any input from the recipient, a scheduled meeting appointment in the calendar folder of the recipient according to the received incoming meeting update.

Abstract: According to examples, an apparatus may include a processor and a memory on which are stored machine-readable instructions that when executed by the processor, may cause the processor to receive an updated data record from a data source and may determine a first conflated data record. The first conflated data record may be associated with the updated data record and include data records in a first grouping from among a plurality of data sources. The processor may identify the data records included in the first conflated data record and may generate a second conflated data record that updates conflations among the updated data record and the identified data records. The second conflated data record may include a second grouping of data records. The processor may replace the first conflated data record with the second conflated data record to incrementally update a set of conflated data records.

Abstract: A method and system for intelligently notifying collaborators of activity on collaborative content include receiving a first indication of a first collaboration activity performed on a first document by a first collaborator among a plurality of first collaborators associated with the first document and responsive to receiving the first indication, obtaining first activity information for the first document comprising the first collaborative activity and one or more second collaborative activities performed on the first document. The method and system determine a lifecycle stage of the first document, analyze the first activity information and the lifecycle stage of the first document to identify one or more second collaborators among the plurality of first collaborators associated with the first document to notify of the first collaborative activity, and provide for displaying the first collaboration activity to the one or more second collaborators.

Abstract: An intelligent agent may assume a ghost presence in a meeting on behalf of a non-participant, monitor data communications between participants of the meeting, and identify appropriate triggers or events relating to the non-participant. For example, the agent may detect questions directed at the non-participant or that the non-participant has the knowledge to answer. The non-participant may be auto-summoned with respect to the meeting when the events relating to the non-participant are identified. The auto-summoning may be performed by communicating information about identified event to the non-participant, for example in real-time. The communication about the identified event may be transmitted over any of a variety of communication channels or modalities, including instant message chat, SMS, phone call, establishing a video call, or pager.

Abstract: Technologies for grouping images, and ranking the images and the groupings, based on entities shown in the images. Images may be grouped based on faces shown in the images. Different images with faces that indicate the same entity (e.g., Adam) may be automatically grouped together. Different images with faces that indicate the same multiple entities (e.g., the people in my family) may also be automatically grouped together. Such automatic grouping may be based on facial recognition technologies. Further, images and groups of images may be automatically ranked based on the faces shown and entities represented. Such rankings may also be influenced by adjacent data that indicates family and friends and the like, and that can be used to identify such entities in the images.

Abstract: One or more digital agents are instantiated to monitor to monitor one or more communication sessions on behalf of a user. Respective digital agents are configured to identify, in the information shared during respective communication sessions, respective one or more subsets of information including one or more items of information relevant to the user. The respective one or more subsets of information identified, by the respective one or more digital agents, are obtained from the one or more digital agents. At least some of the information in the respective subsets of information is presented to the user.