Abstract: Methods and systems are provided for evaluating providing insight into complex machine learning models to facilitate optimization of the models at scale. In particular, a system can guide and/or implement model training and evaluation in various machine learning environments. Further, the system is capable of using standardized evaluation metrics to provide insight into the success of iterations of a machine learning model. Additionally, the system allows for presenting visualizations based on such evaluation metrics to facilitate gaining insight into one or more machine learning models and/or instances of a machine learning model.

Abstract: A method and a system for user identification of a user in a computer system are provided. The method comprising: obtaining a user identifier associated with the user; assigning a respective window identifier to each user interface window opened by the user on the computing device; for a given working session of a pre-determined number of working sessions: storing a sequence of user interface windows opened by the user; identify, within user interface windows opened over the pre-determined number of working sessions, at least one pattern including a pre-determined number of repetitive sequences of user interface windows; generating a set of parameters characterizing a time elapsed between a transition from a first user interface window to an other user interface window within the at least one pattern; using the set of parameters associated with the at least one pattern and the user identifier to train at least one classifier.

Abstract: A graphical user interface described herein can allow a user to easily explore and interact with a complex software application in a distributed computing environment. In one example, a system can generate a graphical user interface (GUI) for display on a display device. The GUI can depict a topological view of a group of software components of a distributed software application in a distributed computing environment. The topological view can include a group of nodes representing the group of software components. Each node in the group of nodes can represent a software component and have a status indicator for at least one container pod including the software component. The topological view can also include at least one link between the group of nodes, where the at least one link represents at least one relationship among the group of software components.

Abstract: Systems and methods for attributing a scroll event are described herein. The system can provide, to a client device, an infinite scroll attribution script. The script can cause the client device to set a dimension of an inline frame, embedded with a content document, of an page to a dimension corresponding to a viewport of an application and determine, responsive to detecting a scroll event, that a first offset between a first content document end and a first viewport end is less than or equal to a first predetermined threshold. The script can further cause the client device to determine, responsive to detecting the scroll event, that a second offset between a second content document end and a second viewport end is greater than or equal to a second threshold and assign the scroll event to the inline frame responsive to the determinations of the first and second offsets.

Abstract: A method for identifying and selecting an object in a virtual or real environment is disclosed. The method comprises: determining a target finder area, at a computing device, within an area displayed by the user interface based on a user profile; determining a list of one or more objects within the target finder area; and presenting one or more of the determined list of one or more objects on the user interface.

Abstract: User interfaces, methods and structures are described for intuitively and fluidly creating new artifacts from existing artifacts and for exploring latent spaces in a visual manner. In example embodiments, source artifacts are displayed along with a selector. The selector is operable to indicate a selected set of the source artifacts by establishing a selection region that includes portions of one or more of the source artifacts displayed. Source vectors are associated with the source artifacts in the selected set. One or more resultant vectors are determined based on the source vectors, and an output artifact is generated based on the one or more resultant vectors.

Abstract: Computer implemented methods and systems are provided for controlling access to one or more computational resources by discriminating between a human user or an automated process. An example system comprises one or more computer processors operating in conjunction with computer memory and data storage to render a set of question images including at least two images having one or more inter-image visually ascertainable relationships, and at least one image having one or more intra-image visually ascertainable relationships, the inter-image visually ascertainable relationships and the intra-image visually ascertainable relationships having one overlapping visually ascertainable relationship.

Abstract: Systems and methods herein provide for capturing and displaying media during an event. One system includes multiple camera devices, each being configured at a location of the event and comprising a global positioning module (GPS) operable to identify the camera device's location. A server is communicatively coupled to a computing device, a display device, and the camera devices through a wireless router using a secure wireless protocol that establishes a secure local area network (LAN) to prevent intrusion to the devices. The server is operable to receive images captured by the camera devices over the LAN, to transfer the images to the computing device for vetting, and to display vetted images on the display device. The server may also control the camera devices and/or provide automatic image processing to the images captured by the camera devices.

Abstract: Methods and systems for enabling organization and control of dashboards, visualizations, and other saved data objects into spaces. An exemplary method includes, based on at least one role of a user, controlling the user's access to a default space and to other spaces of a plurality of spaces, such that the only spaces that the user can access are the default space and the one or more other spaces. Each space can contain a number of saved objects such as dashboards, visualizations, or other objects. The method can provide a graphical user interface for enabling the user to select, as a current space, the default space or one of the other spaces; and in response to the selection, automatically saving new objects generated by the user into the current space; wherein each of the spaces is configured to provide access to certain data objects only or access to certain applications only.

Abstract: A method for generating a chart suggestion list using data present in a data table includes providing the data table, one or more data types of a plurality of cells of the data table and one or more statistics pertaining to the plurality of cells as input to a machine learning model to obtain one or more outputs indicating one or more chart types for the data table, and creating one or more suggested charts based on the indicated chart types. Each of the one or more suggested charts represents at least a subset of the data in the data table. The method also includes generating a chart suggestion list including at least one of the one or more suggested charts.

Abstract: A plurality of rating values corresponding to the performance of each of a plurality of pipelines for a dataset of a plurality datasets is stored. The distance between each pair of datasets within the plurality is determined in a latent space. The distance corresponds to an expected value of a regret incurred when a selected pipeline is rated for a performance of the pipeline for a first dataset. The regret is a monotonically decreasing function of the rating value, according to the pipeline's performance. The plurality of datasets is clustered in the latent space, and a representative dataset in each cluster is identified. In response to receiving a new dataset, a set of preferred pipelines is selected from the plurality, according to a set of rating values corresponding to each of the selected pipelines' performance on the representative dataset. The new dataset is processed in the set of preferred pipelines.

Abstract: A method, computer system, and a computer program product are provided for dynamic viewing of an operation graph depicting a distributed workload in an Information Technology system. A list may be received as input and may include selected nodes. A sub-graph may be abstracted from a global graph. The sub-graph and the global graph may depict interactions among ops objects of the distributed workload using nodes and edges. First nodes may be removed from the global graph according to the list. Direct edges between remaining nodes from the global graph may be maintained. A first indirect edge may be generated based on the removed first nodes. The first indirect edge may run between a first remaining node and a second remaining node. The sub-graph may be presented removed from the global graph.

Abstract: Aspects of the present disclosure provide techniques for improved text classification. Embodiments include providing, based on a text string, one or more first inputs to a summary model. Embodiments include determining, based on one or more first outputs from the summary model in response to the one or more first inputs, a summarized version of the text string. In some embodiments the summarized version of the text string comprises a number of tokens that is less than or equal to a maximum number of input tokens for a machine learning model. Embodiments include providing, based on the summarized version of the text string, one or more second inputs to the machine learning model. Embodiments include determining one or more attributes of the text string based on one or more second outputs received from the machine learning model in response to the one or more second inputs.

Abstract: One embodiment provides a method, including: receiving, within a conversational agent creation framework used to create conversational agents that perform negotiations on behalf of users, (i) a selection of one of a plurality of conversational agent roles (ii) negotiation constraints for the selected role, and (iii) an object for the negotiation; creating a conversational agent (i) having the selected role and (ii) programmed with the negotiation constraints; identifying at least one other conversational agent (i) having an opposing role with respect to the object and (ii) being programmed with opposing role negotiation constraints; and conducting a conversational session between the conversational agent and the at least one other conversational agent, wherein the conversational session comprises the negotiation for the object and wherein the conversational agent and the at least one other conversational agent perform the negotiation in view of the negotiation constraints and the opposing role negotiation con

Abstract: A manufacturing assistance apparatus includes a learning unit and an estimator. The learning unit is configured to load a plurality of pieces of actual measurement data in each of which a gap and a plurality of parameters are associated with each other, and construct an estimation model on the basis of machine learning in which the plurality of pieces of actual measurement data serve as teacher data. The gap is provided between a first workpiece and a second workpiece that eventually structure an airframe of an aircraft and that are eventually fastened to each other. The estimation model estimates the gap from the plurality of parameters. The estimator is configured to derive an estimation value of a length of the gap on which measurement has not yet been performed, on the basis of the estimation model constructed by the learning unit and the plurality of parameters.

Abstract: The invention relates to methods, systems, and computer-readable media related to a system having a plurality of users, designed to display a user-customized subset of item and/or provider information to the user.

Abstract: A method of controlling an environment using a roaming electronic assistant. The method comprises capturing a plurality of associations of voice commands to device commands coded for a first electronic assistant (EA) device by a home electronic assistant (EA) client application, where each association links a voice command to a corresponding device command coded for the first EA device, and where the first EA device takes action in a home environment based on the voice commands, sending the associations of voice commands to device commands coded for the first EA device to an EA client application executing on a computer system, and building a mapping of the associations for the first EA device to associations for a second EA device by the EA client application, where each association for the second EA device links a voice command to a device command coded for the second EA device.

Abstract: The present invention discloses a method, device and medium for structuring a capsule endoscopy report text. The method includes: annotating the report text using an annotation model; storing each named entity classification label in the report text in a hierarchical tree structure according to the annotation information to form a tree structure diagram; parsing the tree structure diagram, extracting abnormal structure data and time parameters, and storing the abnormal structure data and time parameters in a panel data structured manner to form an abnormal structure panel table and a time parameter panel table. The present invention can automatically annotate the capsule endoscopy report through the annotation model, and output parameters of different amount of information in different structures, and quantitative and accurate quality control of the capsule endoscopy process and examination results, which provide sufficient convenience for the electronic medical information of the capsule endoscopy.

Abstract: A proxy server receives from a client device a request for a network resource that is hosted at an origin server for a domain. The request is received at the proxy server as a result of DNS request for the domain returning an IP address of the proxy server instead of an IP address of the origin server. The proxy server retrieves the requested network resource. The proxy server determines that the retrieved network resource includes at least one modification token that is of a type that indicates a threat to the client device. For at least this modification token, the proxy server automatically modifies at least a portion of the retrieved network resource that corresponds to that modification token. The proxy server transmits the modified network resource to the client device.

Abstract: Methods and systems for bound based contextual zoom are disclosed. Legacy WINDOWS desktop and web applications were designed for a mouse or other high-precision pointing device, and as such, small controls and closely grouped controls/icons were commonly designed/used on the screen. Interacting with such applications remoted to a touch-based device (e.g., a smartphone or tablet) is difficult. Bound based contextual zoom overcomes at least some of these limitations by permitting a user to zoom in on an area of interest within an application or desktop, and for a period of time long enough to complete the touch interaction. The area zoomed might be a group of controls on a ribbon bar, the minimize/maximize/close buttons in the top-right corner of a window, a group of form input controls, or any other natural cluster of related user interface (UI) elements.