Abstract: Semantic matching between a source screen or source data and a target screen using semantic artificial intelligence (AI) for robotic process automation (RPA) workflows is disclosed. The source data or source screen and the target screen are selected on a matching interface, semantic matching is performed between the source data/screen and the target screen using an artificial intelligence/machine learning (AI/ML) model, and matching graphical elements and unmatched graphical elements are highlighted, allowing the developer to see which graphical elements match and which do not. The matching interface may also provide a confidence score of the individual matches, provide an overall mapping score, and allow the developer to hide/unhide the matched/unmatched graphical elements. Activities of an RPA workflow may be automatically created based on the semantic mapping that can be executed to perform the automation.

Abstract: Semantic matching between a source screen or source data and a target screen using semantic artificial intelligence (AI) for robotic process automation (RPA) workflows is disclosed. The source data or source screen and the target screen are selected on a matching interface, semantic matching is performed between the source data/screen and the target screen using an artificial intelligence/machine learning (AI/ML) model, and matching graphical elements and unmatched graphical elements are highlighted, allowing the developer to see which graphical elements match and which do not. The matching interface may also provide a confidence score of the individual matches, provide an overall mapping score, and allow the developer to hide/unhide the matched/unmatched graphical elements. Activities of an RPA workflow may be automatically created based on the semantic mapping that can be executed to perform the automation.

Abstract: Semantic matching between a source screen or source data and a target screen using semantic artificial intelligence (AI) for robotic process automation (RPA) workflows is disclosed. The source data or source screen and the target screen are selected on a matching interface, semantic matching is performed between the source data/screen and the target screen using an artificial intelligence/machine learning (AI/ML) model, and matching graphical elements and unmatched graphical elements are highlighted, allowing the developer to see which graphical elements match and which do not. The matching interface may also provide a confidence score of the individual matches, provide an overall mapping score, and allow the developer to hide/unhide the matched/unmatched graphical elements. Activities of an RPA workflow may be automatically created based on the semantic mapping that can be executed to perform the automation.

Abstract: Graphical elements in a user interface (UI) may be detected in robotic process automation (RPA) using convolutional neural networks (CNNs). Such processes may be particularly well-suited for detecting graphical elements that are too small to be detected using conventional techniques. The accuracy of detecting graphical elements (e.g., control objects) may be enhanced by providing neural network-based processing that is robust to changes in various UI factors, such as different resolutions, different operating system (OS) scaling factors, different dots-per-inch (DPI) settings, and changes due to UI customization of applications and websites, for example.

Abstract: Described systems and methods allow the automatic extraction of structured information from images of structured text documents such as invoices and receipts. Some embodiments employ optical character recognition (OCR) technology to extract individual text tokens (e.g., words) and token bounding boxes from a document image. A feature vector of each text token comprises a first part determined according to a character content of the text token, and a second part determined according to an image content of the token's bounding box. A neural network classifier produces a label indicative of a type of information (e.g. “billing address”, “due date”, etc.) carried by each text token. In some embodiments, documents are linearized by ordering text tokens in a sequence according to a reading order of a natural language (e.g., English, Arabic) in which the respective document is formulated. Token feature vectors are fed to the classifier in the order indicated by the token sequence.

Abstract: Documents in a local computing system are parsed for indexing and sent to a cloud-based computing system. The cloud-based search service indexes documents that are parsed on the local computing system, as well as documents that reside in the cloud-based computing system. When a user launches a search, the search is executed against a single search index in the cloud-based computing system, which includes an index of content sources from both the local computing system and the cloud-based computing system.

Abstract: Graphical elements in a user interface (UI) may be detected in robotic process automation (RPA) using convolutional neural networks (CNNs). Such processes may be particularly well-suited for detecting graphical elements that are too small to be detected using conventional techniques. The accuracy of detecting graphical elements (e.g., control objects) may be enhanced by providing neural network-based processing that is robust to changes in various UI factors, such as different resolutions, different operating system (OS) scaling factors, different dots-per-inch (DPI) settings, and changes due to UI customization of applications and websites, for example.

Abstract: Graphical elements in a user interface (UI) may be detected in robotic process automation (RPA) using convolutional neural networks (CNNs). Such processes may be particularly well-suited for detecting graphical elements that are too small to be detected using conventional techniques. The accuracy of detecting graphical elements (e.g., control objects) may be enhanced by providing neural network-based processing that is robust to changes in various UI factors, such as different resolutions, different operating system (OS) scaling factors, different dots-per-inch (DPI) settings, and changes due to UI customization of applications and websites, for example.

Abstract: Graphical elements in a user interface (UI) may be detected in robotic process automation (RPA) using convolutional neural networks (CNNs). Such processes may be particularly well-suited for detecting graphical elements that are too small to be detected using conventional techniques. The accuracy of detecting graphical elements (e.g., control objects) may be enhanced by providing neural network-based processing that is robust to changes in various UI factors, such as different resolutions, different operating system (OS) scaling factors, different dots-per-inch (DPI) settings, and changes due to UI customization of applications and websites, for example.

Abstract: Described systems and methods allow the automatic extraction of structured information from images of structured text documents such as invoices and receipts. Some embodiments employ optical character recognition (OCR) technology to extract individual text tokens (e.g., words) and token bounding boxes from a document image. A feature vector of each text token comprises a first part determined according to a character content of the text token, and a second part determined according to an image content of the token's bounding box. A neural network classifier produces a label indicative of a type of information (e.g. “billing address”, “due date”, etc.) carried by each text token. In some embodiments, documents are linearized by ordering text tokens in a sequence according to a reading order of a natural language (e.g., English, Arabic) in which the respective document is formulated. Token feature vectors are fed to the classifier in the order indicated by the token sequence.

Abstract: Documents in a local computing system are parsed for indexing and sent to a cloud-based computing system. The cloud-based search service indexes documents that are parsed on the local computing system, as well as documents that reside in the cloud-based computing system. When a user launches a search, the search is executed against a single search index in the cloud-based computing system, which includes an index of content sources from both the local computing system and the cloud-based computing system.

Abstract: An alert search mechanism is used with search engines such as a crawler to search for desired documents and/or resources. Particular documents are found by using search queries. The search mechanism track values of a set of relevant properties in queries. Whenever a document is searched for by the system, the values of these set of properties are matched with the old value. If there is no match, this is an indication that the document has changed.

Abstract: A smart algorithm for processing transaction from a crawl queue. If the crawler has in memory a predetermined number of URLs for a given host, the crawler reads from the crawl queue URLs from other hosts. As a result the crawler processes multiple hosts concurrently, and thus, uses machine resources more effectively and efficiently to process the URLs. The smart algorithm can further consider other criteria in deciding which URLs to read from the queue. These criteria can include the response time for each repository (host) the crawler processes. Additionally, the crawler can allocate its resources according to content groups (e.g., two pools), one group for faster content delivery and the second group one for slower content delivery. Thus, crawler resources can be partitioned or divided across different pools depending on repository response time. Other criteria can be provided and considered as well.

Abstract: Balancing resource allocations based on priority may be provided. First, a plurality of repositories may be divided into at least two categories. Next, a first portion of computing resources may be dedicated to a first one of the at least two categories. Then a second portion of the computing resources may be dedicated to a second one of the at least two categories. A crawl may then be performed on the plurality of repositories with the computing resources.

Abstract: A process takes advantage of a structure of a server hosting a network site that includes a change log stored in a database to batch index documents for search queries. The content of the site is batched and shipped in bulk from the server to an indexer. The change log keeps track of the changes to the content of the site. The indexer incrementally requests updates to the index using the change log and batches the changes so that the bandwidth usage and processor overhead costs are reduced.

Abstract: An alert search mechanism is used with search engines such as a crawler to search for desired documents and/or resources. Particular documents are found by using search queries. The search mechanism track values of a set of relevant properties in queries. Whenever a document is searched for by the system, the values of these set of properties are matched with the old value. If there is no match, this is an indication that the document has changed.

Abstract: A process takes advantage of a structure of a server hosting a network site that includes a change log stored in a database to batch index documents for search queries. The content of the site is batched and shipped in bulk from the server to an indexer. The change log keeps track of the changes to the content of the site. The indexer incrementally requests updates to the index using the change log and batches the changes so that the bandwidth usage and processor overhead costs are reduced.