Abstract: An example method includes receiving by a server and from a computing device a natural language-based query, converting the natural language-based query into an embedding formatted query having a vector format, inputting the embedding formatted query into a machine-learning logic model, and determining by the machine-learning logic model a query response to the embedding formatted query that is predicted by the machine-learning logic model to be in context of beauty or cosmetic industry. The machine-learning logic model is trained using a multi-dimensional learned-embedding that includes semantically similar terms in proximity in an embedding space and the embedding space is limited to salient terms associated with the beauty or cosmetic industry. The method also includes communicating by the machine-learning logic model the query response to the computing device.

Abstract: Image data, such as a sequence of video frames, and radar data are received. The image data and the radar data correspond to a scene. Based on the image data, a target object may be detected within the scene and a first probability of the detected target object within the scene is determined. Based on the radar data, the target object may be detected within the scene and a second probability of the detected target object within the scene is determined. Based on the first probability and the second probability, a third probability of the detected target object within the scene is determined. A notification may be provided to at least one device based on the third probability.

Abstract: An example method includes receiving by a server and from a computing device a natural language-based query, converting the natural language-based query into an embedding formatted query having a vector format, inputting the embedding formatted query into a machine-learning logic model, and determining by the machine-learning logic model a query response to the embedding formatted query that is predicted by the machine-learning logic model to be in context of beauty or cosmetic industry. The machine-learning logic model is trained using a multi-dimensional learned-embedding that includes semantically similar terms in proximity in an embedding space and the embedding space is limited to salient terms associated with the beauty or cosmetic industry. The method also includes communicating by the machine-learning logic model the query response to the computing device.

Abstract: An example method includes dividing content into a normal set and an exclusive set, ranking content of the normal set based on a user preference vector, creating a prioritized list of tiers of content, publishing the prioritized list of tiers of content to the network and restricting access to certain tiers of the prioritized list of tiers of content, receiving a selection of a first content from a first tier (that is an unrestricted tier) associated with a first user account, determining a social engagement factor (SEF) for the first user account based on an amount of social media activity over the network associated with the first content and attributed to the first user account, and based on the SEF for the first user account, electronically enabling access to a higher priority tier of the prioritized list of tiers of content for the first user account.

Abstract: A configuration server is provided including one or more processors and a workflow generator executing on the one or more processors. The workflow generator is configured to generate a workflow corresponding to a feature of a service. The configuration server also includes a client user interface configured to receive requests from a client user to enable or disable the feature of the service. A service director is configured to control deployment of the feature of the service by placing the workflow corresponding to the feature in an active state or an inactive state responsive to client requests. Responsive to a determination that the workflow is in the active state, the service director communicates the workflow to a deployment server. The feature associated with the workflow is available to an end user accessing the service when the workflow is communicated to the deployment server.

Abstract: A dynamic group buying system that utilizes a neural network-based machine learning system to generate an optimized price campaign for a product or a product bundle while considering human product seller set bounds consisting of a variable combination of reserve quantity of product, minimum sales price and list price, duration of the campaign, number of levels and maximum price. As an extension of the machine learning capabilities, the system includes a recommendation system that aids the seller to make informed, decisions on factors such as what product(s) bundle should be added to the campaign, what is the ideal time and duration of the campaign, for example. The system is designed to optimize the campaign creation capabilities by analyzing the past product campaigns.

Abstract: A configuration server is provided including one or more processors and a workflow generator executing on the one or more processors. The workflow generator is configured to generate a workflow corresponding to a feature of a service. The configuration server also includes a client user interface configured to receive requests from a client user to enable or disable the feature of the service. A service director is configured to control deployment of the feature of the service by placing the workflow corresponding to the feature in an active state or an inactive state responsive to client requests. Responsive to a determination that the workflow is in the active state, the service director communicates the workflow to a deployment server. The feature associated with the workflow is available to an end user accessing the service when the workflow is communicated to the deployment server.

Abstract: The present disclosure relates generally to a method of executing application code developed using a simplified programming interface. I/O tasks and processing intensive tasks are designated as event callbacks in a single event loop runtime environment. The I/O tasks and processing intensive tasks are generated in association with a user request. Tasks associated with event callbacks are offloaded for execution using a resource based multithreading module to allocate the tasks between a plurality of processors based on the current utilization of each of the plurality of processors and an estimation of computing resources for execution of the tasks. The tasks are executed at the allocated processors of the plurality of processors and the user request is fulfilled using results of the executed tasks.

Abstract: The present disclosure relates generally to systems and methods that include a configuration server with a processing element operable to operate a plurality of containers. A plurality of functionality modules are operable to execute within respective containers of the plurality of containers. Each functionality module includes: an input, a functionality operable to be executed by the processing element to perform an artificial intelligence operation using the input, and an output produced by the functionality. The modular system includes a messaging object operable to connect the output of a first one of the plurality of functionality modules to the input of a second of the plurality of functionality modules. The modular system configures the messaging object to adapt the output of first functionality module to the input of the second functionality module.