11 results about "Web document" patented technology

The present disclosure proposes a method, an apparatus and a computer program product for entity extraction based on edge computing. A web document may be obtained. A text feature of the web document may be identified. A visual feature corresponding to the text feature may be identified. An entity type sequence corresponding to the web document may be extracted based on the text feature and the visual feature.

The application provides a network document relationship extraction method and system, the network document relationship extraction method comprises the following steps: replacing all mentions in each entity with the first occurrence of the mention; training a pre-training model using document-level contrast pre-training and mask language model pre-training respectively; using the trained pre-training model to encode the context of a document sample to obtain a vector representation of each token, converting the named entity recognition information and part-of-speech tagging information of each token into a vector representation respectively and splicing the vector representation of each token to obtain a word embedding representation and a document embedding representation; performing representation enhancement on the entity to obtain a final embedding representation; using a binary cross-entropy loss function to calculate the possibility of the existence of a relationship between entities, and selecting the relationship with the highest possibility as the result of relationship classification. The application can effectively extract the relationship of network documents and provide effective support for network text processing.

The present disclosure is generally directed to methods, apparatus, and computer-readable media (transitory and non-transitory) for learning to automatically navigate interactive web documents and / or websites. More particularly, various approaches are presented for training various deep Q network (DQN) agents to perform various tasks associated with reinforcement learning, including hierarchical reinforcement learning, in challenging web navigation environments with sparse rewards and large state and action spaces. These agents include a web navigation agent that can use learned value function(s) to automatically navigate through interactive web documents, as well as a training agent, referred to herein as a “meta-trainer,” that can be trained to generate synthetic training examples. Some approaches described herein may be implemented when expert demonstrations are available. Other approaches described herein may be implemented when expert demonstrations are not available. In either case, dense, potential-based rewards may be used to augment the training.

An easy search method using a dragged address is proposed, and the method includes in a case of receiving input of a selection on a specific menu from among context menus after a mouse drag of a user, i) automatically requesting a web server linked to the specific menu to search for the address specified by the drag, and ii) automatically displaying a web document received as a response to the automatic request in a new tab or new window of a browser.

Various embodiments include systems and methods to implement processing of web content for vulnerability assessments. A plurality of documents comprising web content may be obtained from multiple different web sources, and the documents may be parsed to determine a set of discrete document chunks. Parsing the documents includes determining whether a document satisfies a segmentation condition for segmenting the document into multiple discrete document chunks using a named-entity recognition system configured to segment the document based at least in part on a vulnerability identification. The discrete document chunks may be stored in a database, where vulnerability information is indexed such that each respective entry in the database corresponds to a respective vulnerability identification and a respective discrete document chunk.

In the present invention, a CRM marketing management system is disclosed. The above CRM marketing management system includes a processing server that transmits a web document toward a member's terminal to render an interface screen in which a plurality of segmented section areas are expressed in a matrix form according to row elements and column elements, wherein each funnel stage from the initial stage to the final purchase stage regarding the product purchase process is represented as a row element and different promotion or campaign types unfolding in each funnel stage are represented as column elements; wherein for each segmented section area, the processing server exclusively displays one of three different states: a state where the development of a promotion or campaign is impossible as a column element to which the corresponding segmented section area belongs under the funnel stage as a row element to which the corresponding segmented section area belongs; a content unloading state where the promotion or campaign cannot be developed as a column element to which the corresponding segmented section area belongs due to the absence of content stored in association with the corresponding segmented section area; and a content loading state where the promotion or campaign can be developed as a column element to which the corresponding segmented section area belongs due to content stored in association with the corresponding segmented section area. The processing server uses customers belonging to each funnel stage as a row element as a population, and belongs to the population Displays the number of customers assigned to different partitioned section areas partitioned according to the column elements, based on the results of a classification task to assign customers to different promotion or campaign types as column elements.

Disclosed are methods and computer systems to generate, update, traverse, and analyze a nodal data structure based on data associated with an entity. The methods and systems disclosed herein describe a server that can generate and link various nodes in a nodal network and parse data into unique data tables. The server then displays a web document having a set of words where each word corresponds to a data table. When a user interacts with a word within the web document, the server identifies a set of nodes associated with the word with which the user has interacted. The server then executes one or more analytical protocols using the identified nodes and displays the results.

The present invention enables easy and efficient synchronization between a document and a database.　A server 1 comprises: an item definition acquisition unit 51; information extraction means such as an information extraction control unit 52 and an extraction model 70; a DB generation unit 53; and an update control unit 54. The item definition acquisition unit 51 acquires a list L that defines items corresponding to information to be extracted from a web document W. The information extraction control unit 52 and the extraction model 70 extract information corresponding to each of the items of the list L from the web document W. The DB generation unit 53 generates an extraction item DB 72 in which the items defined by the list L are associated with the extracted information. When the update of the web document W or the list L has been detected, the update control unit 54 causes the information extraction means to re-execute extraction processing, and updates the extraction item DB 72 on the basis of the result of the re-execution.

The present disclosure is generally directed to methods, apparatus, and computer-readable media (transitory and non-transitory) for leaming to automatically navigate interactive web documents and / or websites. More particularly, various approaches are presented for training various deep Q network (DQN) agents to perform various tasks associated with reinforcement learning, including hierarchical reinforcement learning, in challenging web navigation environments with sparse rewards and large state and action spaces. These agents include a web navigation agent that can use learned value function(s) to automatically navigate through interactive web documents, as well as a training agent, referred to herein as a "meta-trainer," that can be trained to generate synthetic training examples. Some approaches described herein may be implemented when expert demonstrations are available. Other approaches described herein may be implemented when expert demonstrations are not available. In either case, dense, potential-based rewards may be used to augment the training.

A method for enhancing a presentation of a network document by a client terminal with real time social media content. The method comprises analyzing a content in a web document to identify a relation to a first of a plurality of multi participant events documented in an event dataset, each of the plurality of multi participant events is held in a geographical venue which hosts an audience of a plurality of participants, matching a plurality of event indicating lags of each of a plurality of user uploaded media content files with at least one feature of the first multi participant event to identify a group of user uploaded media content files selected from the plurality of user uploaded media content files, and forwarding at least some members of the group to a simultaneous presentation on a browser running on a client terminal and presenting the web document.