Refinitiv US Organization LLC

Thomson Reuters Global Resources

Thomson Reuters Enterprise Centre Gmbh

Thomson Reuters Global Resources Unlimited Company

Thomson Reuters Enterprise Center Gmbh

Jochen Lothar Leidner

Tim Nugent

Sam Chadwick

Ole Siig

Lucas Carstens

Fabio Petroni

Sameena Shah

Mans Olof-Ors

Armineh Nourbakhsh

Jack G Conrad

Blake Howald

Frank Schilder

Peter Pircher

John F Duprey

Trace Liggett

Natraj Raman

Thomas Zielund

Charese Smiley

Aaron Cooper

Jonathan J Medin

Kingsley Martin

Gary Berosik

Daniel Dabney

Vassilis Plachouras

Hiroko Bretz

Vasileios Plachouras

Patrick Hinchin

Stuart Whatley

Zarko Panic

Krzysztof Szymanski

Georgios Gkotsis

Matt Harding

Computer-readable media, systems and methods may improve automatic summarization of transcripts of financial earnings calls. For example, a system may generate segments, such as by disambiguating sentences, from a transcript to be summarized. The system may use an estimator that assesses whether or not the segment should be included in the summary. Different types of estimators may be used. For example, the estimator may be rule-based, trained based on machine-learning techniques, or trained on based on machine-learning with language modeling using natural language processing to fine-tune language models specific to financial earnings calls.

Automatic summarization of financial earnings call transcripts

Computer-readable media, systems and methods may improve classification of content based on a machine-learning natural language processing (ML-NLP) classifier. The system may train a general language model based on a general corpus, further train the general language model based on a domain-specific corpus to generate a domain-specific language model, and conduct supervised machine-learning based on the domain-specific language using topic-specific corpus labeled as relating to topics of interest to generate the ML-NLP classifier. Accordingly, the ML-NLP classifier may be trained on a general corpus, further trained on a domain-specific corpus, and fine-tuned on a topic-specific corpus. In this manner, domain-specific content may be classified into topics of interest. The ML-NLP classifier may classify content into the topics of interest.

Machine-learning natural language processing classifier for content classification

The present invention relates to a computer-based system for identifying supply chain risks and generating supply chain graphs representing an interconnected network of entities. An industrial graph database application is configured to account for direct and indirect (transitive) supplier risk and importance, based on a weighted set of measures: criticality, replaceability, centrality and distance. A graph-based model serves as an interactive and visual supply chain risk and importance explorer. A supply network is induced from textual data by applying text mining techniques to news stories and used to populate the supply chain/graph database.

Risk identification engine and supply chain graph generator

A method of remedying erroneous transactions in a distributed ledger system may include: determining, by a court system associated with the distributed ledger system, a judgment that a transferring, by a contract of the distributed ledger system during a conducting of a financial transaction, of a quantity of tokens of the distributed ledger system, from a first account corresponding to a first counterparty to a second account corresponding to a second counterparty, is erroneous; and in response to determining the judgment, executing, by a court contract of the distributed ledger system, a transfer function, configured to be responsive to the court contract, to transfer at least a portion of the quantity of tokens from the second account to the first account.

Systems and methods for smart contract intervention

A computer implemented method for mining risks includes providing a set of risk-indicating patterns on a computing device; querying a corpus using the computing device to identify a set of potential risks by using a risk-identification-algorithm based, at least in part, on the set of risk-indicating patterns associated with the corpus; comparing the set of potential risks with the risk-indicating patterns to obtain a set of prerequisite risks; generating a signal representative of the set of prerequisite risks; and storing the signal representative of the set of prerequisite risks in an electronic memory. A computing device or system for mining risks includes an electronic memory; and a risk-identification-algorithm based, at least in part, on the set of risk-indicating patterns associated with a corpus stored in the electronic memory.

Method and apparatus for risk mining

A method of providing cross-media event linking may include: receiving, at a first input of an event coreferencing system, a stream of social media postings, and at a second input, a stream of news articles; generating, by the event coreferencing system, a first set of event representations representing events referenced by the social media postings, and a second set of event representations representing events referenced by the news articles; determining, by the event coreferencing system, that at least one of the social media postings references a same event referenced by at least one of the news articles, the determining including determining at least one similarity using data of at least one of the first set of event representations corresponding to the at least one of the social media postings and data of at least one of the second set of event representations corresponding to the at least one of the news articles; and transmitting, by an output of the event resolution system to the user system, an alert including at least one coreferenced event representation representing the event referenced by the at least one of the social media postings and the at least one of the news articles.

Systems and methods for cross-media event detection and coreferencing

Intervention in the operation of a smart contract in a distributed ledger system may include: launching execution of the smart contract; upon launching execution of the smart contract, executing a judgment checking function to determine whether a judgment against the smart contract exists in a judgment database; in response to a judgment against the smart contract existing in the judgment database, intervening in the operation of the smart contract to prevent the execution of the smart contract; and in response to no judgment against the smart contract existing in the judgment database, continuing execution of the smart contract.

Systems and methods for interception of smart contracts

A computer implemented method for mining supply chain information to produce supply chain graphs includes receiving by a computer a set of data; identifying a supplier, a commodity, and a customer from the set of data; generating a query comprising the identified data; determining the absence of any of the supplier, the commodity, or the customer from the set of data; if any items are determined to be absent from the set of data, substituting a placeholder for the missing item from the data set; sending the query; receiving a set of supply chain graph information; generating a supply chain graph signal based upon the set of supply chain graph information; and transmitting the supply chain graph signal. A computing device or system includes a processor an electronic memory; and a program for mining supply chain information to produce supply chain graphs stored in the electronic memory.

Methods and systems for generating supply chain representations

The present invention is directed toward a system for database querying using natural language generation, which comprises identifying a first set of entities corresponding to an indexed data set in response to a user query, generating a ranked list of query intents using the first set of entities, wherein each item of the list of query intents represents a second set of entities associated with the user query and iterating over the ranked list of query intents to identify a top ranked intent associated to one of a set of predefined query plans. The predefined query plan associated with the top rank intent is executed using the set of entities corresponding to the top ranked intent, the predefined query plan comprising one or more search actions against the indexed data set. A first set of results is then received and a description is generated and transmitted.

System for natural language interaction with financial data

A neural paraphrase generator receives a sequence of tuples comprising a source sequence of words, each tuple comprising word data element and structured tag element representing a linguistic attribute about the word data element. An RNN encoder receives a sequence of vectors representing a source sequence of words, and RNN decoder predicts a probability of a target sequence of words representing a target output sentence based on a recurrent state in the decoder. An input composition component includes a word embedding matrix and a tag embedding matrix, and receives and transforms the input sequence of tuples into a sequence of vectors by 1) mapping word data elements to word embedding matrix to generate word vectors, 2) mapping structured tag elements to tag embedding matrix to generate tag vectors, and 3) concatenating word vectors and tag vectors. An output decomposition component outputs a target sequence of tuples representing predicted words and structured tag elements, the probability of each single tuple from the output is predicted based on a recurrent state of the decoder.

Neural paraphrase generator

Exemplary systems for processing a corpus of documents, such as legal contracts or agreements, are disclosed. The systems include a phrase discovery engine which derives statistics and phrase equivalence classes, groups of phrase equivalence classes, and uberphrases (clauses) bounded by phrases. These can be used to determine origins of phrases or clauses within given legal contract or to suggest alternative phrases and clauses.

Systems, methods, and software for questionbased sentiment analysis and summarization

The disclosure relates to entity disambiguation in an interactive fashion based upon attributes. Once disambiguated, a manner in which to review a future article based upon an initial set of attributes and an additional set of attributes and, when needed, generate a warning flag based upon the review is also described. The warning flag may result in a change in the relationship between the entity and an organization such as a bank that is considering engaging in a business relationship with a candidate client, and that may subsequently abandon that plan, or a bank that had previously extended credit to the entity, and that subsequently may be less inclined as a result of the information retrieved by the present invention. A previously accepted entity may also be flagged for credit or anti-money laundering consideration if and when the applicant's situation (e.g., financial situation, involvement in anti-money laundering) changes.

Systems and methods for interactively disambiguating entities

An image of a printed document portion is provided to a synchronizer. The synchronizer retrieves an electronic version of the printed document and identifies an electronic text portion that is textually similar to a printed text portion. The synchronizer detects an annotation in the printed document portion and inserts a corresponding digital annotation into the electronic document.

Growing community of inventors

London, United Kingdom

Jochen Lothar Leidner