Abstract: The present application discloses a method for detecting cigarette appearance defects based on variational Bayesian inference. In a cigarette appearance defect detection scenario, in order to solve the problem that a current point estimation-based machine learning algorithm leads to an overconfidence decision in a data scarcity area, the present application proposes a variational inference-based Bayesian method to improve a SSD backbone network. The present application can improve the accuracy of defect detection without losing the processing speed, not only saving a large amount of manpower and material resources, but also effectively providing reliable technical support for finding the root causes of defect types. Thus, it achieves intelligent quality risk management and control in a cigarette production workshop.

Abstract: A system and method employs a chatbot into which a user can input an inquiry/question regarding a risk related topic is disclosed. The user can obtain an answer to the inquiry. The system and method can receive, via a chatbot interface, an inquiry. An intent of the inquiry can be determined using a natural language processing (NLP) model. The inquiry can be matched, using a classifier, to a related data set based on the determined intent. An output can be generated based on the related data set in response to the inquiry. The output can be transmitted for display on the chatbot interface.

Abstract: Principals of refraction of a light ray are used to identify the location of an interface between two materials having different refractive indices. The interface may be an interface between a plasma layer and a red blood cell layer of a centrifuged container, for example. An array of light detector elements are arranged to receive light that has been refracted through different layers in a centrifuged sample. Elements of the light detector array are arranged at known locations relative to an emitter of the light so detection of light by one or more particular detector elements is indicative of the angle of refraction of the light. Vertical position of the sample is tracked and correlated with corresponding angles of refraction to determine the vertical position of the sample when a change in the angle of diffraction is detected.

Abstract: Systems and methods for preventing blockchain functions associated with malicious entities are disclosed herein. A first blockchain status update may be received from a first authorizing entity. A first counter and a first plurality of authorizing entities can be determined. The first counter and the first plurality of authorizing entities can be updated. The first counter can be compared to a consensus threshold. A first instruction to block requests for communications to the first entity address can be generated. A confirmation of the first instruction can be generated for display.

Abstract: Systems and methods are described for using generative artificial intelligence to automatically label training prompts for training a classification model. Interaction data comprising communications between clients and a chatbot may be used to identify a representative set of topics. Each communication can be assigned a label associated with one or more topics from the set of topics. First training data comprising the set of labels and one or more of the communications assigned to that label may be generated and used to train a generative artificial intelligence model to generate metadata comprising a description of each label from the set of labels. Second training data comprising a plurality of sample prompts, the set of labels, and the metadata; may be generated and used to train a classification model to autonomously label each of the plurality of sample prompts with one or more labels from the set of labels.

Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.

Abstract: Disclosed herein are system, method, and computer program product embodiments for protecting user data privacy in a privacy zone around a user and a computing device displaying user data. A device can include a camera configured to capture a sequence of images in a location, such as in a privacy zone comprising an area around the device and a user of the device. A memory can be coupled to the camera to store the sequence of images. A monitoring device can be configured to determine, based on the sequence of images, whether an object is found in the location. In response to finding an object, the monitoring device can generate a control signal to cause the device to take an action to protect the user data privacy in the location such as a privacy zone.

Abstract: Systems and methods for verifying digitally secured communications between users are disclosed herein. A first user identifier for a first user may be received. A first blockchain address for the first user based on the first user identifier may be determined. A first non-transferable token corresponding to the first user may be generated. A request for a communication may be received. A target address and a target user identifier may be extracted. Whether the communication metadata matches the token metadata may be determined. A validation status for the communication may be determined.

Abstract: Disclosed herein are system, method, and computer program product embodiments for protecting user data privacy in a privacy zone around a user and a computing device displaying user data. A device can include a camera configured to capture a sequence of images in a location, such as in a privacy zone comprising an area around the device and a user of the device. A memory can be coupled to the camera to store the sequence of images. A monitoring device can be configured to determine, based on the sequence of images, whether an object is found in the location. In response to finding an object, the monitoring device can generate a control signal to cause the device to take an action to protect the user data privacy in the location such as a privacy zone.

Abstract: The present application discloses a method for detecting cigarette appearance defects based on variational Bayesian inference. In a cigarette appearance defect detection scenario, in order to solve the problem that a current point estimation-based machine learning algorithm leads to an overconfidence decision in a data scarcity area, the present application proposes a variational inference-based Bayesian method to improve a SSD backbone network, preferably including calculating a real posterior probability by the variational inference method, replacing a point estimation mechanism of a convolution layer weight in the original backbone network by using a form of probability distribution, then detecting the improved extracted features by an SSD target detection algorithm, and furthermore, accurately determining whether the appearance of the cigarette is defective and the type and location of the defect.

Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.

Abstract: In some implementations, a multi-purpose device may include a secure charging location for charging mobile devices. The multi-purpose device may receive authentication information associated with a user. The multi-purpose device may communicate with a fraud detection device, to authenticate the user and to identify that the user is associated with an account. The multi-purpose device may detect that a mobile device associated with the user has been placed in the secure charging location. The multi-purpose device may transmit to the fraud detection device and based on authenticating the user, an indication that the mobile device associated with the user has been placed in the secure charging location.

Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.

Abstract: A computer-implemented method for transmitting electronic data. The method may include receiving a user data set for a user, wherein the user data set includes a user identifier and one or more user authentications for one or more entities, each entity of the one or more entities having an entity eco-friendly score, determining a user eco-friendly score for the user based on the one or more user authentications and a respective one or more entity eco-friendly scores, modifying a data store server to store the user identifier and the eco-friendly score for the user, generating a targeted event associated with a user medium, based on the user eco-friendly score, and transmitting, to a user device of the user, the targeted event via the user medium.

Abstract: Systems and methods for preventing blockchain functions associated with malicious entities are disclosed herein. A first blockchain status update may be received from a first authorizing entity. A first counter and a first plurality of authorizing entities can be determined. The first counter and the first plurality of authorizing entities can be updated. The first counter can be compared to a consensus threshold. A first instruction to block requests for communications to the first entity address can be generated. A confirmation of the first instruction can be generated for display.

Abstract: Systems and methods for verifying digitally secured communications between users are disclosed herein. A first user identifier for a first user may be received. A first blockchain address for the first user based on the first user identifier may be determined. A first non-transferable token corresponding to the first user may be generated. A request for a communication may be received. A target address and a target user identifier may be extracted. Whether the communication metadata matches the token metadata may be determined. A validation status for the communication may be determined.

Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.

Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.

Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.

Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.