Abstract: In an approach to improve improving impact prevention in both manual and autonomous vehicles embodiments of the present invention generate a knowledge corpus based on data collected, from one or more internet of thing (IoT) sensors, associated with one or more predetermined risk factors of one or more predefined risks over time. Further, embodiments receive from the one or more IoT sensors input data to observe a vertical space above a vehicle and identify an object in a predetermined area based on the received input data and the knowledge corpus. Additionally, utilize the knowledge corpus and the input data to generate a risk score for the identified object and responsive to the risk score being above a predetermined threshold, embodiments issue a command for the vehicle to continue to drive or to progress to a predetermined safe zone.

Abstract: Generation of control instructions to control operation of conveyor includes receiving image data associated with each section of a plurality of sections of a conveyor. Each section of the plurality of sections includes at least one adjustable supporting structure. The operational data associated with the conveyor is received. An artificial intelligence (AI) model is applied on the image data and the operational data. Anomaly data is determined for an anomaly associated with the conveyor. A set of control instructions are generated based on the anomaly data. The set of control instructions is associated with an operation of at the conveyor. The set of control instructions is outputted.

Abstract: The present disclosure provides a processing method of a remote attestation report, a database service end and a database client end. The method includes: acquiring a request statement related to a remote attestation report from a database client end; executing a preset function corresponding to the request statement to call an interface of a first TEE remote attestation SDK to obtain a processing result related to the remote attestation report; sending the processing result to the database client end. Embodiments of the present disclosure can reduce the difficulty in the development and application of a remote attestation function.

Abstract: In an omnichannel call center, a voice call between a caller and an agent is monitored, including generating text of the voice call in real time, and using Natural language processing (NLP) on the text to determine whether the call has been prematurely disconnected. If the call was prematurely disconnected, an alternative communication channel between the caller and the agent is established.

Abstract: A computer-implemented method for dynamically determining a database (DB) isolation level of a database generates a transition matrix correlating a set of database transaction types with inter-arrival times, transaction service times, and isolation levels. A probability of success for combinations of transaction type and isolation levels is provided based on the inter-arrival times and transaction service times. Recommended database isolation adjustments are determined based on the probabilities of success. A current database isolation level of the database is adjusted using a database manager based on the recommended database isolation adjustment.

Abstract: The program intelligently assigns a virtual office layout (VOL) to one or more actual users' real on-site office layout (ROL) based on human resource mapping. The method maps the one or more actual users' ROL to a correlated VOL, according to human resource data and a real building floor map. The method creates an augmented virtual office layout (AVOL) based on the mapping. The method customizes the AVOL according to the one or more actual users' preferences and real office location and tracks the one or more users' activities on the AVOL. The method renders the one or more users' real or preferred office location and the one or more users' virtual position on the AVOL.

Abstract: While organizations can employ IPsec based VPNs to securely connect different sites (e.g., branch sites, data centers, and/or virtual private clouds), the security can disrupt network performance by obfuscating information used for load balancing. Disclosed is technology that employs minimal decryption in a secure manner to load balance multiple network traffic flows within a secure connection (“tunnel”) across security appliances that effectively operate as alternative endpoints for the tunnel. The security appliances within a load balancing pool are configured/programmed to share tunnel keys with each other after tunnel establishment and with the load balancer. The load balancer uses the tunnel keys to minimally decrypt in a lookaside memory encrypted packets to ascertain N-tuples, The load balancer then uses the N-tuples to load balance the flows within a tunnel across the security appliances.

Abstract: A method for identifying a target region of a digital pathology slide, including: obtaining a scanned image of a pathology slide; inputting the scanned image of the pathology slide into a preset deep learning-based identification model; extracting a contour feature of the scanned image of the pathology slide by using an image contour feature extraction submodel, to obtain a contour image; segmenting the contour image by using an image segmentation submodel to obtain a plurality of sub-contour images; separately performing classification and identification on the plurality of sub-contour images by using an image classification submodel, to obtain a region category corresponding to each sub-contour image; and determining a target region image based on the region category of each sub-contour image. In addition, a system for identifying a target region of a digital pathology slide, a device, and a medium are further proposed.

Abstract: In an approach to autonomous vehicle traffic management, one or more computer processors retrieve an identification associated with a first autonomous vehicle. One or more computer processors retrieve a distance rule corresponding to the first autonomous vehicle from a smart contract, wherein the distance rule is a rule for a minimum distance that other autonomous vehicles must maintain with respect to the first autonomous vehicle. One or more computer processors determine a context of the first autonomous vehicle. Based on the context of the first autonomous vehicle, one or more computer processors determine whether a first exception to the distance rule is required. One or more computer processors generate the first exception to the distance rule. One or more computer processors communicate the distance rule and the first exception to the distance rule to one or more additional autonomous vehicles traveling in proximity to the first autonomous vehicle.

Abstract: Embodiments determine at least one product interest based on user information, receive historical data from a plurality of users, train a convolutional neural network (CNN) model based on the historical data, determine a plurality of user task interactions related to the at least one product interest based on the trained CNN model, monitor the user task interactions for a user product related to the at least one product interest over a predetermined period of time, generate a digital twin of a comparative product to the user product based on the monitored user interactions, generate user performance metrics including a return on investment (ROI) of the generated digital twin, and generate a recommendation to purchase a product of the at least one product interest based on the generated user performance metrics including the ROI being greater than a cost of the product.

Abstract: While organizations can employ IPsec based VPNs to securely connect different sites (e.g., branch sites, data centers, and/or virtual private clouds), the security can disrupt network performance by obfuscating information used for load balancing. Disclosed is technology that employs minimal decryption in a secure manner to load balance multiple network traffic flows within a secure connection (“tunnel”) across security appliances that effectively operate as alternative endpoints for the tunnel. The security appliances within a load balancing pool are configured/programmed to share tunnel keys with each other after tunnel establishment and with the load balancer. The load balancer uses the tunnel keys to minimally decrypt in a lookaside memory encrypted packets to ascertain N-tuples. The load balancer then uses the N-tuples to load balance the flows within a tunnel across the security appliances.

Abstract: A method for providing context-aware cueing for daily interactions, navigation, and accessibility is disclosed. In one embodiment, such a method includes receiving routine data describing routine-based tasks associated with a user. The method further receives interaction data describing current interactions of the user in a particular environment. The method processes, using a neural network, the routine data and interaction data to determine a next step to be performed by the user that is consistent with the routine-based tasks and the current interactions. The method provides a cue to the user to perform the next step in the particular environment. This cue may include one or more of a visual, auditory, and haptic cue. The method may continuously refine the neural network to increase the accuracy of the cue. A corresponding system and computer program product are also disclosed.

Abstract: A method for tuning software container interdependent systems includes receiving, by a computer processor operating a prediction engine, data from one or more software containers. The data includes operations from one or more software applications in the software containers. The prediction engine identifies a plurality of dependencies in the operations. The prediction engine develops a model of interdependencies based on the identified dependencies. The prediction engine determines one or more failure candidates in the identified dependencies. The prediction engine generates a software architecture template based on the identification of failure candidates.

Abstract: Mitigating GPS inaccuracy is provided. The method comprises identifying a required level of location accuracy for an activity of an autonomous system and determining, according to GPS coordinates, a location of the autonomous system. A determination is made, according to a knowledge base created by historical learning, whether GPS location accuracy at the location of meets the required level of location accuracy for the activity. Responsive to a determination that the GPS location accuracy at the location of the autonomous system does not meet the required level of accuracy, location determination is handed off to an installed infrastructure within proximity of the autonomous system which provides the required level of location accuracy. The autonomous system then performs the activity according to location information provided by the installed infrastructure.

Abstract: Methods for intelligent management of asset workflows are disclosed herein. One method includes predicting that a user is going to perform a predetermined action based on one or more user movements included in a signal received from a wearable sensing device and generating, by a processor, a warning of one or more consequences of the user performing the predetermined action. The method further includes displaying, on augmented reality glass, the warning to the user prior to the user actually performing the predetermined action. Also disclosed herein are systems and apparatus that can include, perform, and/or implement the operations of the methods.

Abstract: A system for layer-by-layer three-dimensional (3D) printing of a 3D printed object. The system includes a printing component configured to print a plurality of layers to form the 3D printed object, a surface modification component configured to be co-operable with the printing component and configured to apply an appropriate degree of surface roughness to at least one of the plurality of layers after being printed, and a controller configured to identify parameters of each of the at least one of the plurality of layers and configured to control operation of the printing component and the surface modification component, wherein the parameters include the appropriate degree of surface roughness.

Abstract: Provided is a method, system, and computer program product for generating automated guidance based on situational analysis. A processor may receive a set of actions that a user requires assistance performing. The processor may identify IoT devices associated with the user. The processor may analyze real-time data from the IoT devices to determine a contextual surrounding of the user. The processor may determine that a first contextual surrounding matches a first action of the set of actions that the user requires assistance performing. The processor may determine, based on analyzing a current state of the user from the real-time data, if the user is in an optimal state to receive guidance. The processor may generate, in response to the user being in the optimal state, guidance for assisting the user to perform the first action. The processor may provide the guidance to the user via the IoT devices.

Abstract: The disclosure provides a lithium ion battery module which includes: a plurality of cores, being winded cores, laminated cores or pouch batteries, one end of each core being provided with a positive tab and a negative tab, the plurality of cores being connected in series and/or in parallel by using positive tabs and negative tabs through connecting sheets, the positive tabs and negative tabs forming a total positive electrode; a plurality of partition members, each partition member being used for accommodating a single core and physically separating respective cores, the partition members having a top opened structure; and a housing and a cover plate, configured to form an inner space after being assembled for accommodating partition members and cores, the cover plate being provided with a connecting part for a total module positive electrode and a connecting part for a total module negative electrode.

Abstract: An embodiment for improved authentication of chatbot users for chat history recovery. The embodiment may record a first chat conversation between a first chat service operator and a first user, wherein the first chat conversation includes a first-step authentication and a chat history. The embodiment may detect a request to resume the first chat conversation. The embodiment may, in response to detecting the request to resume the first chat conversation, initiate a recovery process by sending a random-generated token to the first-step authentication. The embodiment may receive verification of the random-generated token. The embodiment may generate, by a large language model, one or more authentication questions based on content of the chat history. The embodiment may authenticate the first user based on one or more correct answers to the one or more authentication questions. The embodiment may resume the first chat conversation including the chat history.

Abstract: An approach is provided for partnership optimization. Using a supervised machine learning model, a categorized profile of partners is generated based on feedback from clients and past performances of the partners. The categorized profile indicates strengths of the partners. A network graph is generated based on data about interactions between the partners and clients and the categorized profile. The network graph has nodes representing the partners and the clients and edges representing connections between the partners and the clients. Using the categorized profile and the network graph, a three-dimensional model is generated and represented by a three-dimensional matrix of cells. A given cell represents a part of a project and includes constraint(s) of the project. Using a machine learning algorithm based on Thistlethwaite's algorithm, the model is solved to optimally match the constraint(s) with strength(s) of partner(s) and strength(s) of connection(s) between the partner(s) and client(s).