In the field of government engagement management, for users of an employee desktop web client, it is now possible, within the web client application, to search and read articles and/or knowledge content that has been authored to external locations. Due to this integration to external, third-party applications, content and/or articles can be displayed to an agent on the employee desktop web client graphical user interface. Agents can enter free text into a specific search field and review the results in summary form, and then select an article in HTML format to progress the current interaction with the client. This functionality adds value to the agent experience and enables the agent to provide an improved service to the end client. Results may be filtered by the search engine as well. Moreover, this system and method improves the operation of the computer in that the computer running such a system in the past was not able to integrate in such a fashion in a web client format. This system and method also enables an agent to handle calls with the web client more efficiently, and allows agents on the web client to automatically classify.

Examples of the present disclosure describe content creation of at least one abstract from extracted content. Input information is received that comprises a link to a landing page and keyword information. The landing page is evaluated and content including at least one of image data and text data is extracted from the landing page. Properties of the content extracted are analyzed. The content extracted is ranked based on the analyzed properties and application of at least two ranking algorithms. A first ranking algorithm applied ranks the content based on relevance to the landing page and a second ranking algorithm applied ranks the content extracted based on relevance to the keyword information. The ranked content is filtered to remove content or portions of content that are determined to be unappealing based on applying filtering rules to the ranked content. At least one abstract is created from the filtered and ranked content.

Methods and systems for generating and using a conversation summary model. The method comprises receiving at least one training dataset. The at least one training dataset comprises data samples, each data sample comprising a text comprising text segments. The text is labelled with a conversation summary comprising any of the text segments which summarize the text. The at least one training dataset includes a dataset from a specific source. Using the at least one training dataset and the pre-trained model, the method further comprises generating the conversation summary model by fine-tuning the pre-trained model. The generated conversation summary model may be used to generate conversation summaries for chat conversations.

Examples disclosed herein relate to segmentation based on clustering engines applied to summaries. In one implementation, a processor segments text based on a comparison of the output of multiple clustering engines applied to multiple summarizations of documents associated with the text. The processor outputs information related to the contents of the segments.

A system and method for summarizing suggested content and sharing the summarized suggested content is described. In one aspect, a computer-implemented method includes performing an analysis of text of a document, searching a document library for content elements and documents based on the analysis of the text, identifying candidate documents and candidate content based on the searching, presenting a list of candidate documents or candidate content with the document authoring application, receiving a selection of a candidate document or candidate content from the list in the document authoring application, and providing the selected candidate document to a collaborative content sharing platform, the collaborative content sharing platform configured to generate a graphical user interface that displays a list of shared documents, the shared documents includes candidate documents selected by one or more users of a group of users that share access to the collaborative content sharing platform.

A method for generating summaries includes selecting a first subset of text units of a text composition to incorporate into a first summary of the text composition using a weighting of the text units that indicates for each text unit a relative importance of including the text unit in summaries of the text composition. The weighting of the text units is modified to reduce the relative importance of each text unit in the first subset based on the text unit having been selected for the first subset. A second subset of the text units is selected to incorporate into a second summary of the text composition using the modified weighting of the text units. At least one of the first summary and the second summary are provided to a user device.

A method may include obtaining a document and using a first prediction model to generate text block scores for text blocks in the document, where a first text block of the text blocks is associated with a first text block score of the plurality of text block scores. The method also includes updating, in response to the first text block score for the first text block failing to satisfy a criterion, a modified version of the document with an indicator to set the first text block as a hidden text block in a presentation of the modified version. The method also includes generating a summarization of the first text block based on the words in the first text block and updating the modified version of the document to include the summarization. The method also includes providing the modified version of the document to a user device.

A source content routing system is described for distributing source content received from clients such as documents, to translators for performing translation services on the source content. The routing system extracts source content features, which may be represented as vectors. The vectors may be assembled into an input matrix, which may be processed using an artificial neural network, classifier, perceptron, CRF model, and/or the like, to select a translator such as a machine translation system and/or human. The translator provides translation services translation from a source language to a target language, post translation editing, proof reading, quality analysis of a machine, quality analysis of human translation, and/or the like and returns the product to the content routing system or clients.

The present disclosure provides a summary generation model training method and apparatus, a device and a storage medium, and relates to the field of computer technologies, and in particular, to the field of artificial intelligence such as natural language processing and deep learning. The summary generation model training method includes: acquiring a document representation corresponding to a document sample; constructing, based on the document representation, a summary representation corresponding to the document representation, the summary representation including a positive summary representation and a negative summary representation; and constructing a total contrastive loss function based on the document representation, the positive summary representation and the negative summary representation, and training a summary generation model based on the total contrastive loss function. The present disclosure may improve accuracy of the summary generation model.

An apparatus and computer implemented method that include obtaining, into a computer, text of a patent, automatically finding and extracting, using the computer, a set of claim text from the patent text, identifying, using the computer, text of independent claims from the set of claim text, displaying in a first row on a computer monitor the text of the independent claims, automatically determining a plurality of preliminary scope-concept phrases from the text of the independent claims, displaying in a second row on the computer monitor the text of the plurality of preliminary scope-concept phrases, eliciting and receiving user input to specify a first one of the plurality of preliminary scope-concepts phrases, and highlighting each occurrence of the specified first one of the plurality of preliminary scope-concept phrases in a plurality of the independent claims displayed in the first row. A scope concept builder tool is also provided.