A computer-implemented method for performing quality review of life science documents is described. One or more of the life science documents are scanned by a mobile device, wherein the one or more life science documents are sent to a database. Language, image, rotation, and noise are among the content that is checked among the life science documents, and wherein similarities, suspicious changes, document layouts, and missing sections are checked among the one or more life science documents. In addition, feedback is sent by a system to an originator of the life science documents based on the content regarding imaging, rotation, and noise and the similarities, suspicious changes, document layouts and missing sections.

A vehicle number identification device includes a registration-number-information acquisition part configured to acquire registration number information associated with an identification media installed in a vehicle; a number-plate-image acquisition part configured to acquire a number-plate image of the vehicle; an OCR processing part configured to acquire OCR resultant information representing a result of an optical character recognition with respect to the number-plate image; an estimation processing part configured to input the number-plate image into a machine-learning model and to acquire estimation result information output from the machine-learning model; and a matching determination part configured to determine whether at least two types of information out of the registration number information, the OCR resultant information, and the estimation result information indicate a same vehicle number and to identify the same vehicle number as a vehicle number of the vehicle.

A computing system generates a training data set for training the prediction model to detect defects present in a target surface of a target specimen and training the prediction model to detect defects present in the target surface of the target specimen based on the training data set. The computing system generates the training data set by identifying a set of images for training the prediction model, the set of images comprising a first subset of images. A deep learning network generates a second subset of images for subsequent labelling based on the set of images comprising the first subset of images. The deep learning network generates a third subset of images for labelling based on the set of images comprising the first subset of images and the labeled second subset of images. The computing system continues the process until a threshold number of labeled images is generated.

A character recognition system converting image data of a character string into a character code is provided, which includes a data processing unit, and a storage unit configured to be used by the data processing unit, wherein the storage unit stores similar-outline nonsense character string data that has an outline similar to that of to an ordinary character string but does not make sense, and the data processing unit performs character recognition processing using the similar-outline nonsense character string data that is stored in the storage unit.

Systems and methods are provided for an improved machine learning (ML) model system. The improved ML system can be configured to (1) initially classify the types of images and videos received by the various devices and provide the classified input to different ML models based on the classification (e.g., of the distortion level, etc.), and/or (2) reuse portions (referred to as base components) of each ML model where parameters of the base components are unchanged across the various ML models, while replacing other portions (referred to as adapted components) of the ML model where the parameters of the adapted components may change greatly.

A system for performing document extraction is configured to: (a) receive a first document; (b) extract the first document into document elements, the document elements including pages, lines, paragraphs, or any combination thereof; (c) determine a first set of fields of interest for the first document, wherein the first set of fields of interest are determined via a type of the first document or via a first set of queries for probing the first document; (d) determine, from a plurality of closed domain question answering (CDQA) models, a first set of CDQA models that provides answers to each field of interest included in the first set of fields of interest; and (e) provide answers to the first set of fields of interest to the client device.

Examples of techniques for interactive generation of labeled data and training instances are provided. According to one or more embodiments of the present invention, a computer-implemented method for interactive generation of labeled data and training instances includes presenting, by the processing device, control labeling options to a user. The method further includes selecting, by a user, one or more of the presented control labeling options. The method further includes selecting, by a processing device, a representative set of unlabeled data samples based at least in part on the control labeling options selected by the user. The method further includes generating, by a processing device, a set of suggested labels for each of the unlabeled data samples.

A classification model training method is provided. The method includes selecting a training dataset, determining a category of a sketch in the training dataset according to a sketch classification model, to obtain a first category processing result, and analyzing, according to a second feature analysis model, a feature of the sketch extracted by a first feature extracting model, to obtain a second analysis result of the sketch; then obtaining a function value of a first loss function according to the first category processing result and the second analysis result of the sketch; and finally adjusting a first model parameter value of the sketch classification model according to the function value of the first loss function.

An object-extraction method includes generating multiple partition objects based on an electronic document, and receiving a first user selection of a data element via a user interface of a compute device. In response to the first user selection, and using a machine learning model, a first subset of partition objects from the multiple partition objects is detected and displayed via the user interface. A user interaction, via the user interface, with one of the partition objects is detected, and in response, a weight of the machine learning model is modified, to produce a modified machine learning model. A second user selection of the data element is received via the user interface, and in response and using the modified machine learning model, a second subset of partition objects from the multiple partition objects is detected and displayed via the user interface, the second subset different from the first subset.

Systems and methods for automatic labeling of data with user validation and/or correction of the labels. In one implementation, unlabeled images are received at an execution module and changes are made to the unlabeled images based on the execution module's training. The resulting labeled images are then sent to a user for validation of the changes. The feedback from the user is then used in further training the execution module to further refine its behaviour when applying changes to unlabeled images. To train the execution module, training data sets of images with changes manually applied by users are used. The execution module thus learns to apply the changes to unlabeled images. The feedback from the user works to improve the resulting labeled images from the execution module.