Refinitiv US Organization LLC

Thomson Reuters Enterprise Centre Gmbh

Financial & Risk Organisation Limited

Fabio Petroni

Jochen Lothar Leidner

Sameena Shah

Armineh Nourbakhsh

John F Duprey

Tim Nugent

Natraj Raman

Lucas Carstens

Zarko Panic

Vasileios Plachouras

Vassilis Plachouras

Jochen Leidner

John Duprey

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

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

Growing community of inventors

London, United Kingdom

Fabio Petroni