Abstract: A detecting deepfake system for detecting deepfake in live network traffic communications and stored media. Visual data and audio data are extracted from a plurality of users of an enterprise. Digital fingerprints are generated from the extracted visual data and audio data. The digital fingerprints are matched with the pre-stored fingerprints to produce an output. Further, the visual data and audio data are analyzed by Artificial Intelligence (AI)-detection model. A deepfake detection score is generated based on the analysis by the AI-detection model. The output from the matching of the digital fingerprints with the pre-stored fingerprints and the deepfake detection score are combined to determine a level of deepfake in the extracted visual data and audio data. The policies are applied based on the level of deepfake detected in the visual data and the audio data.

Abstract: A cloud-based network security system (NSS) is described. The NSS uses a sandbox to safely open and extract information about a PDF file and uses machine learning algorithms to analyze the information to predict whether the PDF file contains malware. Specifically, dynamic information about the PDF file is captured while it is open in the sandbox. Static information is extracted from the PDF file as well. The dynamic and static information is input to an AI or machine learning model trained to provide an output indicating a prediction of whether the PDF file contains malware. A verdict engine uses the output from the AI or machine learning model to classify the document as malicious or clean. Security policies can then be applied based on the classification.

Abstract: A cloud-based network security system (NSS) is described. The NSS uses a sandbox to safely open and extract information about a PDF file and uses machine learning algorithms to analyze the information to predict whether the PDF file contains malware. Specifically, dynamic information about the PDF file is captured while it is open in the sandbox. Static information is extracted from the PDF file as well. The dynamic and static information is input to an AI or machine learning model trained to provide an output indicating a prediction of whether the PDF file contains malware. A verdict engine uses the output from the AI or machine learning model to classify the document as malicious or clean. Security policies can then be applied based on the classification.

Abstract: A cloud-based network security system (NSS) is described. The NSS uses a sandbox to safely open and extract information about a PDF file and uses machine learning algorithms to analyze the information to predict whether the PDF file contains malware. Specifically, dynamic information about the PDF file is captured while it is open in the sandbox. Static information is extracted from the PDF file as well. The dynamic and static information is input to an AI or machine learning model trained to provide an output indicating a prediction of whether the PDF file contains malware. A verdict engine uses the output from the AI or machine learning model to classify the document as malicious or clean. Security policies can then be applied based on the classification.

Abstract: The technology discloses training a classifier to label webpages with categories, extracting from a training database with thousands of webpages tentatively labeled with ground truth categories, dataset A and dataset B; training a classifier using A; and applying it to webpages in B to assign a webpage a label and a classification score. Also disclosed is cleaning B, removing webpages based on evaluation of a decision confidence metric derived from the score assigned for the webpage; and training a second classifier using cleaned B, with second classifier weights initialized independent of trained first classifier weights; and applying the second classifier to webpages in A to assign a webpage the label, score, and decision confidence matrix, and cleaning A, removing webpages from A based on the decision confidence metric. Then combining cleaned A and cleaned B into a combined clean dataset, and training the third classifier using the combined clean dataset.

Abstract: Disclosed are a sample analysis apparatus and method which include: obtaining a first light signal by irradiating a first test sample from blood and a first-channel reagent to differentiate neutrophils, eosinophils, lymphocytes, and monocytes; obtaining a second light signal from a second test sample from the blood and a second-channel reagent; counting the nucleated red blood cells and lymphocytes based on the second light signal; determining accuracy of the lymphocyte result from the first test sample; and if inaccurate, correcting the lymphocyte result based on the lymphocyte result of the second test sample.

Abstract: High-density socket connectors are disclosed herein. An example socket connector includes a first housing to be coupled to a substrate, The first housing has a first side, a second side opposite the first side, and a first opening extending through the first housing between the first side and the second side. The socket connector includes a socket pin disposed in the first opening. The socket connector further includes a second housing moveably coupled to the first housing. The second housing has a third side, a fourth side opposite the third side, and a second opening extending through the second housing between the third side and the fourth side. The socket pin extends into the second opening of the second housing. The second housing is moveable relative to the first housing between a first position and a second position.

Abstract: A cloud-based network security system (NSS) is described. The NSS uses a sandbox to safely detonate and extract information about a document and uses machine learning algorithms to analyze the information to predict whether the document contains malicious software. Specifically, during the detonation, static and dynamic information about the document is captured in the sandbox as well as character strings from images in the document. The dynamic information (and sometimes the static information) is input to an AI or machine learning model trained to provide an output indicating a prediction of whether the document contains malware. The character strings are compared with a batch of phishing keywords to generate a heuristic score. A validation engine combines the output from the AI or machine learning model and the heuristic score to classify the document as malicious or clean. Security policies can then be applied based on the classification.

Abstract: Disclosed technology of training a third classifier to select between sensitive or non-sensitive categories for a webpage including both sensitive and non-sensitive content. The technology involves collecting webpages as a tentatively labeled dataset A and a dataset B, training a first classifier including at least a first sensitive category classifier and a first non-sensitive category classifier using the tentatively labeled dataset A, referring some dual labelled webpages generated by the first classifier to a curator to curate and resolve label conflict, receiving curated labels from the curator and updating dataset B with the curated labels, training a second classifier using the updated dataset B, referring some dual labelled webpages generated by the second classifier to a curator to curate and resolve label conflict, receiving curated labels from the curator and updating dataset A with the curated labels, and training the third classifier with updated dataset A and updated dataset B.

Abstract: The technology discloses training a classifier to label webpages with categories, extracting from a training database with thousands of webpages tentatively labeled with ground truth categories, dataset A and dataset B; training a classifier using A; and applying it to webpages in B to assign a webpage a label and a classification score. Also disclosed is cleaning B, removing webpages based on evaluation of a decision confidence metric derived from the score assigned for the webpage; and training a second classifier using cleaned B, with second classifier weights initialized independent of trained first classifier weights; and applying the second classifier to webpages in A to assign a webpage the label, score, and decision confidence matrix, and cleaning A, removing webpages from A based on the decision confidence metric. Then combining cleaned A and cleaned B into a combined clean dataset, and training the third classifier using the combined clean dataset.

Abstract: A cloud-based network security system (NSS) is described. The NSS uses a sandbox to safely detonate and extract information about a document and uses machine learning algorithms to analyze the information to predict whether the document contains malicious software. Specifically, during the detonation, static and dynamic information about the document is captured in the sandbox as well as character strings from images in the document. The dynamic information (and sometimes the static information) is input to an AI or machine learning model trained to provide an output indicating a prediction of whether the document contains malware. The character strings are compared with a batch of phishing keywords to generate a heuristic score. A validation engine combines the output from the AI or machine learning model and the heuristic score to classify the document as malicious or clean. Security policies can then be applied based on the classification.

Abstract: A semiconductor chip package is described. The semiconductor chip package has a substrate. The substrate has side I/Os on the additional surface area of the substrate. The side I/Os are coupled to I/Os of a semiconductor chip within the semiconductor chip package. A cooling assembly has also been described. The cooling assembly has a passageway to guide a cable to connect to a semiconductor chip's side I/Os that are located between a base of a cooling mass and an electronic circuit board that is between a bolster plate and a back plate and that is coupled to second I/Os of the semiconductor chip through a socket that the semiconductor chip's package is plugged into.

Abstract: An apparatus and method for reducing differential cross-talk in a CPU pin arrangement of a server motherboard are described. The CPU pin arrangement has pin patterns that are arranged in a square, separated from each other, and are mirrored around an axis between the pin patterns. Each pin pattern has pins that each include a main body with a lateral asymmetric cross-section, a first connector extending from one end of the main body and coupled to the motherboard through solder, and a second connector extending from an opposing end of the main body and coupled to the CPU through a pressure contact.

Abstract: A sample analysis apparatus and a sample analysis method. The method includes preparing a specimen by means of a sample and a reagent, the sample being a blood sample or a body fluid sample. The specimen is tested to obtain one or more testing parameters of the sample with respect to particles; and for at least one testing parameter, control is performed according to the magnitude of the testing parameter to display the testing parameter with different precisions. By performing the control according to the magnitude of a testing parameter to display the testing parameter with different precisions, a result requirement of a low-value sample, such as a body fluid sample, can be satisfied.

Abstract: Provided are a sample analysis apparatus and method. The method includes: determining particle test items in a particle test item set of a sample, e.g., blood or body fluid, to obtain test results of the particle test items; obtaining a sample display mode, which includes a first display mode and a second display mode, where the first display mode displays a test result of a particle test item, which belongs to a first test item set, and the second display mode displays a test result of a particle test item, which belongs to a second test item set.

Abstract: A sample analyzer and an animal reticulocyte testing method. After collecting a blood sample of an animal, the sample analyzer can determine, according to the animal type of the animal, a reticulocyte incubation time corresponding to the animal, and control an incubation time of the blood sample on the basis of the reticulocyte incubation time to prepare a sample fluid to be tested. The fragmentation of reticulocytes in said sample liquid obtained by using the reticulocyte incubation time is within a preset reasonable range, such that the accuracy of a testing result of the reticulocytes is ensured, and the application range of the animal testing is improved.

Abstract: Disclosed are a sample analysis apparatus and a sample analysis method. The method includes: obtaining a leukocyte measurement mode for a test sample, the method having a plurality of leukocyte measurement modes, each of which includes at least a reaction time, and different leukocyte measurement modes having different reaction times; determining a reaction time for the current sample to be tested; controlling the sample to be tested to react with the reagent, to prepare a test sample for detecting leukocytes; and controlling to detect the test sample for detecting leukocytes, to obtain detection data of leukocyte classification.

Abstract: Disclosed is technology for choosing between alternative category labels tentatively assigned to tens of thousands of webpages by a classifier ensemble running on processors, applying the classifier ensemble with a sensitive category classifier, a non-sensitive category classifier, a title and metadata classifier and a heuristic classifier to tens of thousands of webpages. Also disclosed is applying a post processor to outputs of the classifier ensemble and tentatively assigning at least two category labels for non-sensitive categories for webpages; two category labels, automatically determining that at least one but not all of the tentatively assigned category labels is a general label and de-assigning the general label; saving the category label that is not de-selected to the webpage; and distributing the assigned category labels for at least some of the tens of thousands of webpages for use in controlling access to webpages by users on user systems protected using the assigned labels.

Abstract: A method, computer system, and computer program product are provided for mitigating network risk. A plurality of risk reports corresponding to a plurality of network devices in a network are processed to determine a multidimensional risk score for the network. The plurality of risk reports are analyzed using a semantic analysis model to identify one or more factors that contribute to the multidimensional risk score. One or more actions are determined using a trained learning model to mitigate one or more dimensions of the multidimensional risk score. The outcomes of applying the one or more actions are presented to a user to indicate an effect of each of the one or more actions on the multidimensional risk score for the network.

Abstract: A method, computer system, and computer program product are provided for network risk analysis. A plurality of risk reports relating to a network device in a network are obtained, wherein each risk report is associated with a particular dimension of a plurality of dimensions of risk for the network device in the network. A count of the plurality of risk reports is determined for each dimension of the plurality of dimensions of risk. A regression model is applied to determine a risk value for the network device in the network based on the count of the plurality of risk reports for each dimension and based a role of the network device in the network.