Abstract: A computer implemented method comprising receiving invoice data comprising at least one of invoice descriptions and invoice characters from user computers, each of the invoice descriptions and invoice characters defines contexts and patterns, wherein each of the invoice data comprising a supplier-customer information that corresponds to a supplier-customer transaction; analyzing the at least one of the invoice descriptions and the invoice characters with corresponding contexts and patterns; determining that amount of the invoice characters is more than a threshold number of characters, for performing: matching invoice data, invoice characters with predefined historical invoice data that corresponds to the same supplier-customer information; computing a similarity score for each of the invoice data; and displaying recommendations including first account codes to map the one or more first account codes to each of the one or more invoice data based on the similarity score; determining that amount of the invoice

Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput. Embodiments are capable of converting a table portion of a read-only document to a searchable, editable data record using text rectangle (TR)-level numerical data that indicates probabilities of TRs belonging to canonicals and at least one convolutional neural network (CNN) that processes the TR-level numerical data to produce table-level numerical data.

Abstract: Disclosed herein is a method of generating a 3D conformation of a molecule used in in-silico drug discovery. The method includes obtaining a 2D graph of the molecule. A first tensor including molecular graph information and a second tensor including graph, coordinate and distance information of atoms of the molecule are generated. A first set of feature vectors corresponding to the first tensor and a second set of feature vectors corresponding to the second tensor are further generated. The first set of feature vectors are fed to a first encoder of a generative model to generate a first output. The second set of feature vectors are fed to a second encoder of the generative model to generate a second output. The first output and the second output are combined to form an input in a decoder of the generative model to generate the 3D conformation of the molecule.

Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput. Embodiments are capable of converting a table portion of a read-only document to a searchable, editable data record using text rectangle (TR)-level numerical data that indicates probabilities of TRs belonging to canonicals and at least one convolutional neural network (CNN) that processes the TR-level numerical data to produce table-level numerical data.

Abstract: A frequency conversion power transmission system includes: a new energy power generation base, a first isolation device, a second isolation device, an alternating current-alternating current (AC-AC) frequency conversion device and a power transmission cable; the new energy power generation base is configured to supply electrical energy to an AC power grid, and operate at a constant voltage and a constant or variable frequency according to environmental conditions including weather, an environment or a distance; the first isolation device is connected to the new energy power generation base; the second isolation device is connected to the AC power grid; an input terminal of the AC-AC frequency conversion device is connected to the first isolation device, an output terminal of the AC-AC frequency conversion device is connected to the second isolation device, and the power transmission cable is configured to connect the new energy power generation base and the first isolation device.

Abstract: Disclosed herein is an automated small molecule generation process for use in in silico drug discovery. The automated process employs a trained neural network that analyzes a graph adjacency tensor which represents a small molecule compound. Over subsequent iterations, the trained neural network analyzes the graph adjacency tensor and predicts actions (e.g., adding an atom, adding a bond type, or assigning a charge) that, if taken, are likely to lead to a valid small molecule compound. Thus, the methods described herein generate small molecule compounds of increased validity in comparison to conventional methodologies.

Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput.

Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput. Embodiments are capable of converting a table portion of a read-only document to a searchable, editable data record using text rectangle (TR)-level numerical data that indicates probabilities of TRs belonging to canonicals and at least one convolutional neural network (CNN) that processes the TR-level numerical data to produce table-level numerical data.

Abstract: Described herein is a computer-implemented method for automatic extraction of canonical data from an electronic document. The method comprises classifying a first text rectangle in an electronic document as a label and a second text rectangle as a value using a first machine learning algorithm. A first probability score of a likelihood of the first text rectangle corresponding to a first canonical category is determined using a second machine learning algorithm. A second probability score of a likelihood of the second text rectangle corresponding to a first canonical category is determined using a third machine learning algorithm. A relative spatial position of the second text rectangle relative to the first text rectangle is calculated. Based on the relative spatial position, the first probability score, and the second probability score, the first text rectangle, and the second text rectangle are classified into the first canonical category.

Abstract: The present disclosure provides a light bar assembly, a backlight unit, a display unit and a mobile electronic device. The light bar assembly includes: a light bar printed circuit board; a plurality of backlights electrically connected to the light bar printed circuit board; and at least one till light electrically connected to the light bar printed circuit hoard. The light bar printed circuit board includes a first circuit and a second circuit connected to a drive circuit, the first circuit is configured to provide an electrical signal to the backlights, and the second circuit is configured to provide an electrical signal to the fill light.

Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput. Embodiments are capable of converting a table portion of a read-only document to a searchable, editable data record using text rectangle (TR)-level numerical data that indicates probabilities of TRs belonging to canonicals and at least one convolutional neural network (CNN) that processes the TR-level numerical data to produce table-level numerical data.

Abstract: A dual camera assembly, an electronic apparatus and a method of acquiring an image are provided. The dual camera assembly includes: a first camera lens and a second camera lens; a first sensor configured for receiving light that has passed through the first camera lens; a liquid crystal light valve and a polarizer which are on a side, which is close to the first camera lens, of the first sensor. The polarizer is on a side, which is close to the first sensor, of the liquid crystal light valve, and liquid crystal molecules in the liquid crystal light valve are rotatable.

Abstract: A detection device is provided and the detection device includes: a body, including a first groove and a first opening; the first opening is on a first surface of the body and in communication with the first groove; a collector, configured to be at least partially accommodated into the first groove in a rotatable manner and being capable of rotating to allow at least part of the collector to extend out of the first opening in a process of detection; a test paper supplier, disposed on the first surface and configured to provide a test paper on a side of the first opening away from the first groove in the process of detection so as to collect a detection sample; and a detector, disposed on the first surface and configured to detect the test paper that has collected the detection sample.

Abstract: The present disclosure relates to a seat, comprising: a seat base; a seat frame; and a sitting posture detection device configured to alert a user when a sitting posture of the user sitting on the seat deviates from a standard sitting posture, wherein the sitting posture detection device includes: one or more distance sensors configured to measure a distance from the seat frame to one or more body parts of the user; and a processor configured to determine, according to the distance detected by the distance sensor, whether the sitting posture of the user deviates from the standard sitting posture, and issue an alarm information if the sitting posture of the user deviates from the standard sitting posture.

Abstract: The present disclosure provides a light bar assembly, a backlight unit, a display unit and a mobile electronic device. The light bar assembly includes: a light bar printed circuit board; a plurality of backlights electrically connected to the light bar printed circuit board; and at least one till light electrically connected to the light bar printed circuit hoard. The light bar printed circuit board includes a first circuit and a second circuit connected to a drive circuit, the first circuit is configured to provide an electrical signal to the backlights, and the second circuit is configured to provide an electrical signal to the fill light.

Abstract: A detection device is disclosed and the detection device includes: a body, including a first groove and a first opening; the first opening is on a first surface of the body and in communication with the first groove; a collector, configured to be at least partially accommodated into the first groove in a rotatable manner and being capable of rotating to allow at least part of the collector to extend out of the first opening in a process of detection; a test paper supplier, disposed on the first surface and configured to provide a test paper on a side of the first opening away from the first groove in the process of detection so as to collect a detection sample; and a detector, disposed on the first surface and configured to detect the test paper that has collected the detection sample.

Abstract: The present disclosure relates to a seat, comprising: a seat base; a seat frame; and a sitting posture detection device configured to alert a user when a sitting posture of the user sitting on the seat deviates from a standard sitting posture, wherein the sitting posture detection device includes: one or more distance sensors configured to measure a distance from the seat frame to one or more body parts of the user; and a processor configured to determine, according to the distance detected by the distance sensor, whether the sitting posture of the user deviates from the standard sitting posture, and issue an alarm information if the sitting posture of the user deviates from the standard sitting posture.

Abstract: A dual camera assembly, an electronic apparatus and a method of acquiring an image are provided. The dual camera assembly includes: a first camera lens and a second camera lens; a first layer configured for receiving light that has passed through the first camera lens; a liquid crystal light valve and a polarizer which are on a side, which is close to the first camera lens, of (he first sensor. The polarizer is on a side, which is close to (he first sensor, of the liquid crystal light valve, and liquid crystal molecules in the liquid crystal light valve are rotatable.

Abstract: Embodiments of the present disclosure provide a knee joint brace and a knee joint supporting method using the same. The knee joint brace includes: a support frame including two support stands, the two support stands being arranged in parallel with and spaced from each other; at least one adjusting band connected to the support frame; a rotating dual-disc disposed on each of the two support stands, each of the two support stands being capable of bending about the rotating dual-disc; and a supporting band disposed at one end of the support frame and configured to support the support frame in an extending direction of the support stand.

Abstract: A handheld electronic device is provided. The handheld electronic device includes a housing, wherein a universal ball structure is provided at a corner of the housing. The universal ball structure includes a groove, and a ball in the groove.