Abstract: Among other things, embodiments of the present disclosure improve the functionality of electronic messaging and imaging software and systems by generating and displaying media overlays with avatars of different users. For example, media overlays can be generated by the system and displayed in conjunction with media content (e.g., images and/or video) generated by an image-capturing device (e.g., a digital camera).

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for augmenting neural networks with an external memory. One of the systems includes a memory interface subsystem that is configured to perform operations comprising determining a respective content-based weight for each of a plurality of locations in an external memory; determining a respective allocation weight for each of the plurality of locations in the external memory; determining a respective final writing weight for each of the plurality of locations in the external memory from the respective content-based weight for the location and the respective allocation weight for the location; and writing data defined by the write vector to the external memory in accordance with the final writing weights.

Abstract: A website building system (WBS) includes a processor implementing a machine learning feedback-based proposal module and a database storing at least the web sites of a plurality of users of the WBS, and components of the websites. The module includes a plurality of per activity AI units and a feedback system. Each per activity AI unit supports one or more specific activity related to the WBS and provides at least one system suggestion to the users related to its specific activity. Each per activity AI unit includes at least one machine learning model suitable for the activity supported by its per activity AI unit. The feedback system provides a plurality of different kinds of feedback from the users and from rule engines for updating the machine learning models. The feedback system analyzes the feedback to determine which one of the at least one machine learning models to update.

Abstract: An unstructured document analysis method according to an embodiment includes: operations of acquiring unstructured document data including font characteristic data and document structure data, extracting text included in the unstructured document data on the basis of the font characteristic data or the document structure data, classifying the extracted text into a pre-classified item using a trained neural network model, acquiring a content query related to the content included in the unstructured document data and associated with the pre-classified item, and generating an answer to the content query on the basis of the extracted text classified into the item.

Abstract: The present disclosure is directed towards systems and methods for extracting structure and headers from a body of text. This computational extraction is based on the visual and logical similarities between portions of text. Structure is derived from a programmatic and methodic computation of similarities between header pairs.

Abstract: In various embodiments, the disclosed systems and methods may receive documents, analyze the documents, categorize portions of the analyzed documents, and present the images of the documents and at least a portion of the categories. The analysis may include identification of categories and the presentation may include indicia of the portion of the image of the document related to the category. The systems and methods disclosed may allow querying and/or reporting of a plurality of documents to facilitate processing.

Abstract: Provided is a method for automatically creating a user-customized document, and the method includes: receiving an automatic document creation target from a user, wherein the automatic document creation object comprises a plurality of data items and a plurality of tags matching the plurality of data items is defined; receiving, from the user, a template file in which a format of the user-customized document is set, wherein the template file comprises at least one tag; and generating the user-customized document by replacing each tag in the template file with a data item matching a corresponding tag in the automatic document creation target.

Abstract: Systems and methods for replacing a hyperlink to a referenced object with a smart link are disclosed. The method performed by a smart link management system includes receiving a content request for the hyperlink from a client device, the content request comprising a uniform resource locator (URL) of the referenced object, the URL indicating the location of the content; identifying a data source where the referenced object is located based on the URL; forwarding the content request to the data source; receiving metadata corresponding to the referenced object hosted by the data source; and forwarding the metadata to the client device for display on the client device in the form of a smart link, the smart link being a visual object including at least a portion of the metadata received from the data source.

Abstract: Disclosed herein are embodiments of systems, methods, and products comprises a server for database search and knowledge mining. The server may learn different table's semantics, relationships, and usage by parsing historical query logs and analyzing tables' metadata (e.g., table descriptions). The analytic server may generate a graph database based on the table relationships obtained from the parsing. The graph database may be a relationship graph where tables are the nodes and edges represent the relationships among tables. When the server receives a query, the server extract semantics of the query, and return a set of tables that are semantically similar to the query. The set of tables may be a list of tables whose semantic similarities with the query satisfies a threshold. The analytic server may further generate a graph including the list of tables to show the relationships of these tables.

Abstract: Embodiments relate to a document parser module that can receive a document template from a producing entity, parse the template, and generate a mapping between the template and identifiers of the transaction entity. Using the mapping and parsed information, a document generator module may generate a document that includes the required identifier data and meets the submission standards of the producing entity. A configuration file may be saved so that a document can be generated and submitted every reporting period with little or no additional modification by the transaction entity.

Abstract: A user device to receive, from an infrastructure device, a trained machine learning (ML) model to enable the user device to determine a given type of input information to be auto-fill in an observed field portion in an observed network element; analyze an observed source code associated with the observed network element to determine an observed characteristic associated with the observed field portion that is configured to accept the given type of input information; calculate an observed signature associated with the observed field portion based at least in part on the observed characteristic; utilize the trained ML model to evaluate the observed signature to determine the given type of input information; and auto-fill, in the observed field portion, input information in accordance with the given type of input information is disclosed. Various other aspects are contemplated.

Abstract: The object is to provide a control device, an operation control device, a server, a management server, a recording medium, a neural network system model, a control method and an operation control method that enable reuse of combustible waste as industrial raw materials with high efficiency. A control device comprises a gas information acquisition unit that acquires gas information on gas converted by a gasifying furnace for converting collected waste to gas; a control information acquisition unit that acquires control information controlling the gas purification device for purifying gas converted by the gasifying furnace; a feature information acquisition unit that acquires feature information including information on purified gas purified by the gas purification device; and a creation unit that creates a learning model by machine learning based on the gas information, the control information and the feature information.

Abstract: Identification of an electronic communication containing specific information is provided. Content of the electronic communication may be evaluated by a machine-learning model, and based on an evaluation of the content, it may be determined that the electronic communication contains the specific information. The electronic communication may be tagged with tag information indicating that the electronic communication contains the specific information, and transmission of the electronic communication may be blocked based on the tag information.

Abstract: Embodiments are disclosed for recommending fonts based on text inputs are described. In some embodiments, a method of recommending fonts includes receiving a selection of text, providing a representation of the selection of text to a font recommendation model, generating, by the font recommendation model, a prediction score for each of a plurality of fonts based on the representation of the selection of text, and returning at least one recommended font based on the prediction score for each of the plurality of fonts.

Abstract: Methods, systems, and computer-readable medium are disclosed for presenting web application content. For example, a computer-readable medium may include instructions executable by one or more processors to perform operations including: executing a parent application using a web browser, the parent application being a single page application configured to display a web page element in the web browser; identifying a child application requested by a user action; obtaining, from a server system, child application code of the identified child application, the child application code defining a micro-frontend in a first programming language; transpiling the child application code from the first programming language to a second programming language executable by the web browser, to obtain a transpiled child application code renderable by the web browser; and rendering the transpiled child application code so that the micro-frontend is displayed in the web browser.

Abstract: An apparatus, computer-implemented method, and a system are disclosed that maintain a knowledge base to pre-populate prospective database inputs, by: receiving one or more database input forms containing database input fields; associating the database input fields to previous database input fields stored in a repository based on a recognition process on corresponding respective word strings and previous word strings corresponding to the previous database input fields; retrieving prior inputs to the previous database input fields based on the associating; formatting the retrieved one or more prior inputs; updating the received one or more database input forms; transmitting the updated database input forms; receiving at least one of a confirmation and an edit; finalizing the updated one or more database input forms; and transmitting the finalized database input forms to one or more source computing apparatuses associated with the one or more database input forms.

Abstract: In variants, a method for asset generation can include: determining campaign settings, determining recipient data, determining a digital asset template, mapping variables between the digital asset template and the recipient data, optionally generating a set of example digital assets, rendering a set of digital assets, optionally printing a physical version of each digital asset, optionally delivering the physical assets. However, the method can additionally or alternatively include any other suitable elements.

Abstract: In a method for securing a web browser, display instructions for displaying web content are received from a content server by a web browser operating on a client computing device. The web browser creates a document object using the display instructions and determines from the document object whether one or more of the display instructions meet font profiling criteria selected to identify attempts to profile font display characteristics of the web browser. Responsive to a determination that font profiling criteria are met, the web browser alters a character display of the document object and implements the document object for displaying the web content on the client computing device.

Abstract: A new implementation is provided for an error back-propagation algorithm that is suited to the hardware constraints of a device implementing a spiking neural network. The invention notably uses binary or ternary encoding of the errors calculated in the back-propagation phase to adapt its implementation to the constraints of the network, and thus to avoid having to use floating-point number multiplication operators. More generally, the invention proposes a global adaptation of the back-propagation algorithm to the specific constraints of a spiking neural network. In particular, the invention makes it possible to use the same propagation infrastructure to propagate the data and to back-propagate the errors in the training phase. The invention proposes a generic implementation of a spiking neuron that is suitable for implementing any type of spiking neural network, in particular convolutional networks.

Abstract: Systems and methods for linking a data set to respective locations of relevant data elements in documents included in electronic files in which a formatting server identifies a specific location of each occurrence of each of a plurality of different categories of fields in documents in an electronic file and links an identifier of each of the categories to each specific location of each of the occurrences. Thereafter, a review server compares data populating all of the occurrences at each specific location of at least one of the categories of fields to which the identifier of the particular category of fields is linked and generates a notification when the data populating any of the occurrences differs from data populating any of the other occurrences of the same category of fields.