Abstract: A system determines an exposure of datasets to changes in constraints associated with data analytics, for example, policy constraints of an organization. The system stores a set of policy constraints associated with datasets of an organization, each constraint describing a step performed for creating a dataset. The system generates execution plans for natural language data analytics requests. The execution plan includes steps, each step comprising a set of instructions. One or more steps of the execution plan are associated with a constraint. The system generates the dataset by executing the execution plan and stores an association between the dataset and the constraint. The system uses the associations to identify datasets and execution plans that may be impacted as a result of modification of a policy constraint and generates a report to determine whether the modification to the policy constraint be permitted.

Abstract: A system performs in-context learning of data analytics agents by generating execution plans for natural language data analytics requests specific to the data agent. A user provides several natural language data analytics requests relevant to the context for which the data analytics agents are being trained. Multiple users may provide feedback during planning phase, execution phase of the execution plans of the natural language data analytics requests and result analysis phase. Explicit feedback provided by users as well as implicit feedback is used for evaluating the new execution plan to determine whether the new execution plan should replace an existing execution plan stored in the system for a matching natural language data analytics request.

Abstract: A system performs verification of results of natural language data analytics requests using a machine learning based language model. The system generates an execution plan comprising ordered data processing steps to produce an analytical result. The system traverses the execution plan in reverse, and for each step, identifies applicable invariants from a seed library or by generating new ones, produces structured query input describing the step and invariants, and uses the language model to generate executable verification code. The verification code is executed on step inputs/outputs to determine invariant satisfaction without exposing raw data to the model. Discrepancies are localized to specific steps, correlated with possible root causes, and presented in a detailed verification report with lineage and metrics.

Abstract: A system generates partially materialized execution plans for processing the natural language data analytics requests. The partially materialized execution plan comprises at least a set of instructions identified as materialized. A materialized set of instructions remains unchanged if the partially materialized execution plan is regenerated from the natural language data analytics request using the machine learning based language model. The system iteratively refines the partially materialized execution plan, by performing the following steps repeatedly. The system stores a finalized execution plan obtained by iteratively refining the partially materialized execution plan and executes the finalized execution plan to generate a report that may be displayed via a user interface, for example, a dashboard.

Abstract: A system allows multiple users to interact with the system to perform data analysis. The system allows users to specify data analytics requests using high-level natural language requests. The system generates an execution plan based on natural language requests received from a particular user and executes the execution plan. The system detects a failure in executing the execution plan and determines whether to request help from another entity. The system may change the state of execution of the execution plan to a waiting state in which the system blocks execution until it receives an approval to proceed from an entity. The system receives a request to modify the execution plan from the entity and modifies the execution plan based on the request. The system iteratively performs the modification of the execution plan until an acceptable execution plan is generated.

Abstract: A system determines an exposure of datasets to changes in constraints associated with data analytics, for example, policy constraints of an organization. The system stores a set of policy constraints associated with datasets of an organization, each constraint describing a step performed for creating a dataset. The system generates execution plans for natural language data analytics requests. The execution plan includes steps, each step comprising a set of instructions. One or more steps of the execution plan are associated with a constraint. The system generates the dataset by executing the execution plan and stores an association between the dataset and the constraint. The system uses the associations to identify datasets and execution plans that may be impacted as a result of modification of a policy constraint and generates a report to determine whether the modification to the policy constraint be permitted.

Abstract: A technique for generating a three-dimensional (3D) image using a plurality of terminals is provided. The technique includes performing calibration of a distance between lenses in a plurality of terminals by using view images of a two-dimensional (2D) object acquired by the plurality of terminals, acquiring view images of a 3D object while maintaining a lens distance between the plurality of terminals which is determined based on the result of the calibration, and processing the acquired view images of the 3D object in order to generate a 3D image of the 3D object. The 3D image allows a user to experience the same sense of depth as viewing the object in the real world.

Abstract: A method of controlling a display apparatus using a camera based device and a mobile device, display apparatus, and system applying the same are provided. The method for controlling the display apparatus with the mobile device includes capturing an entity displayed which is on the display apparatus, in response to a movement of the mobile device, sensing at least one of a movement and a change of size of the captured entity, generating a control signal for controlling the display apparatus based on at least one of the sensed movement, and change of size of the entity and transmitting the generated control signal to the display apparatus.

Abstract: A technique for generating a three-dimensional (3D) image using a plurality of terminals is provided. The technique includes performing calibration of a distance between lenses in a plurality of terminals by using view images of a two-dimensional (2D) object acquired by the plurality of terminals, acquiring view images of a 3D object while maintaining a lens distance between the plurality of terminals which is determined based on the result of the calibration, and processing the acquired view images of the 3D object in order to generate a 3D image of the 3D object. The 3D image allows a user to experience the same sense of depth as viewing the object in the real world.

Abstract: A method of controlling a display apparatus using a camera based device and a mobile device, display apparatus, and system applying the same are provided. The method for controlling the display apparatus with the mobile device includes capturing an entity displayed which is on the display apparatus, in response to a movement of the mobile device, sensing at least one of a movement and a change of size of the captured entity, generating a control signal for controlling the display apparatus based on at least one of the sensed movement, and change of size of the entity and transmitting the generated control signal to the display apparatus.