This document describes techniques and devices for a refined search with machine learning. These techniques improve computer-aided searches through enabling selection of search criteria used in a prior search and providing a refined search result based on that selection. Furthermore, a machine-learning component of a search engine can be altered to improve future search results based on the selection and an indication of the desirability of the refined search result.

Disclosed is a method for suggesting revisions to a document-under-analysis (“DUA”) from a seed database, the seed database including a plurality of original texts each respectively associated with one of a plurality of final texts. The method includes tokenizing the DUA into a plurality of statements-under-analysis (“SUAs”), selecting a first SUA of the plurality of SUAs, generating a first similarity score for each of the plurality of the original texts, the similarity score representing a degree of similarity between the first SUA and each of the original texts, selecting a first candidate original text of the plurality of the original texts, and creating an edited SUA (“ESUA”) by modifying a copy of the first SUA consistent with a first candidate final text associated with the first candidate original text.

Cloud storage provides accessible interfaces, near-instant elasticity, scalability, multi-tenancy, and metered resources in a distributed framework providing fault tolerant solutions with high data durability. Stored data may have legal or compliance requirements defining retention periods ensuring the data is preserved without modification for a period of time. However, data privacy rules such as the European Union's General Data Protection Regulation can require modification or destruction of records at any point. Further, many retention structures are user driven but users make mistakes requiring a change to the record's associated retention period. Retention period mechanism enforced with immutable storage can satisfy compliance requirements but run contrary to data privacy rules as well as blocking adjustments. Accordingly, processes, methods and systems are required allowing retention policy application to data being stored within network based storage as well as allowing retention policies to be applied to stored data thereby facilitating retention period adjustments.

Aspects of the subject disclosure may include, for example, a service platform that receives a chat request from a client device associated with a user. A chat database of previous chat sessions associated with other users is searched based on the chat request to identify a previous chat session corresponding to the chat request. Chat data corresponding to the previous chat session is retrieved from the chat database and sent to the client device. Other embodiments are disclosed.

The present invention provides a seeded news event clustering and retrieval system configured to first create a candidate data set of documents, second create a set of initial clusters based on nearness or duplicate similarity status, and third create an aggregate cluster by merging initial clusters with seed documents. The invention generates top-level clusters for news events based on an editorially supplied topical label or “seed” component and generates sub-topic-focused clusters based on algorithm. The system uses an agglomerative clustering algorithm to gather and structure documents into distinct result sets. Decisions on whether to merge related documents or clusters are made according to similarity of evidence derived from two distinct sources, one, relying on a digital signature based on the unstructured text in the document, the other based on the presence of named entity tags that have been assigned to the document by an event or named entity tagger such as the Thomson Reuters Calais engine/web service.

An algorithm download method, a device, and a related product. The method includes: obtaining an algorithm identifier of an algorithm and a capability description of a client; sending the algorithm identifier and the capability description to a cloud; and receiving a version code that is of the algorithm and that is returned by the cloud, where the version code is obtained by the cloud by searching based on the algorithm identifier and the capability description. According to the foregoing solution, an algorithm can be easily downloaded.

The present disclosure generally relates to the connection, operation, automation, and autonomation of disparate computing systems, networks, and databases. The present disclosure further relates to a method for operating a plurality of legacy systems, comprising: receiving a user input string from a user; parsing the user input string into a plurality of terms; generating, for each of the plurality of legacy systems, based on the user input string, one or more legacy input strings, each of the one or more legacy input strings being executable by each of the plurality of legacy systems and comprising a syntactical arrangement of one or more of the plurality of terms; and instructing the plurality of legacy systems to execute the one or more legacy input strings.

Training and/or utilizing a machine learning model to generate request agnostic predicted interaction scores for electronic communications, and to utilization of request agnostic predicted interaction scores in determining whether, and/or how, to provide corresponding electronic communications to a client device in response to a request. A request agnostic predicted interaction score for an electronic communication provides an indication of quality of the communication, and is generated independent of corresponding request(s) for which it is utilized. In many implementations, a request agnostic predicted interaction score for an electronic communication is generated “offline” relative to corresponding request(s) for which it is utilized, and is pre-indexed with (or otherwise assigned to) the electronic communication. This enables fast and efficient retrieval, and utilization, of the request agnostic interaction score by computing device(s), when the electronic communication is responsive to a request.

Techniques are disclosed that enable learning an embedding space using cross-examples, where a distance between a query and an electronic resource in the embedding space provides an indication of the relevance of the electronic resource to the query. Various implementations include learning the embedding space using cross-example Softmax techniques. Various implementations include leaning the embedding space using cross-example negative mining. Additional or alternative techniques are disclosed that enable determining an electronic resource for a query based on comparing a query vector (e.g., a embedding space representation of the query) with a set of pre-stored candidate electronic resource vectors (e.g., an embedding space representation of a set of candidate electronic resources).

A document creation system includes keyword object management configured to create keyword objects by merging the expression labels and attributes of keywords and to manage the keyword objects, keyword object placement configured to place, in any position of text, a keyword object managed by the keyword object management, keyword object attribute value input configured to input attribute values to the keyword objects, and keyword object-placed document recording/reading configured to record and read the created keyword object-placed document.