Gregory Grefenstette

At least since the invention of writing, people have been troubled by the problem of what a word means. Dictionaries have traditionally been written with numbered word senses, giving the impression that the different senses of a word are fixed abstract entities, which can be used to separate usages into neat piles according to their different meanings. Adam Kilgarriff’s daring 1997 article ‘I Don’t Believe in Word Senses’ challenged this traditional view of word meaning, presenting an account in which ‘the basic units are occurrences of the word in context’. Kilgarriff went on to develop the Sketch Engine, a statistical tool that enables lexicographers and NLP researchers, teachers, and students to explore the relationship between meanings and collocations (words and their contexts). In this review article, we compare the information provided by Kilgarriff’s Word Sketches with the recently developed Word Embedding techniques and with the results of Corpus Pattern Analysis.

It is easy to find large-scale, free-to-use, computer readable geographical databases nowadays, and many people are examining how to exploit them, crossing them with other linked data to produce new applications. We present here a publicly accessible web application for searching for movies based on location, mashing up information from DBpedia and GeoNames. We also describe its architecture and the solutions we developed to address the entity ambiguity problems we encountered.

Beyond the theoretical linguistics search for language universals, there exist a number of practical linguistic problems concerning multilingual environments. This article discusses a number of related multilingual natural language processing (NLP) applications, many connected with using Internet language identification. Internet users want to shop, interact in virtual communities, and find information without language being a barrier, with as much ease as they can. This article presents a technical overview of the research and development being performed on four multilingual applications: language identification, terminology alignment, cross-language information retrieval, and comprehension aids. It covers a wide range of tools that allows the user to understand text written in another language, without using the machinery of full automatic translation.

Video search is still largely based on text search human-supplied metadata, sometimes supplemented by extracted thumbnails. We have been developing a broadcast news search system based on recent progress in automatic speech recognition (ASR), natural language processing (NLP) and video and image processing to provide a rich content-based search to news. Our public online demonstrator of the Voxalead application, described here, currently indexes daily broadcast news content from 60 sources in English, French, Chinese, Arabic, Spanish, Dutch, Italian, German and Russian and make them searchable few time after they have been published.

One important class of online videos is news broadcasts. Most news organizations provide immediate access to topical news broadcasts over the Internet, through RSS streams or podcasts. Until lately, technology has not made it possible for a user to automatically find, within a longer broadcast, the smaller parts that might interest them. Recent advances in both speech recognition systems and natural language processing have led to a number of robust tools that allow us to provide users with quicker and more focused access to relevant segments of one or more news broadcast videos. Here we present an interface for browsing or searching news broadcasts (video or audio) that exploits these new language processing tools to (i) provide immediate access to topical passages within news broadcasts, (ii) browse news broadcasts by events as well as by people, places and organizations, (iii) search for news through a map interface, (iv) browse news by trending topics, and (v) have a brief automatically generated textual characterization of news segments before listening. Our publicly searchable demonstrator currently indexes daily broadcast news content from sources in English, French, Chinese, Arabic, Spanish, Dutch, Russian and Italian. This demonstrator can be consulted online at: http://vox.labs.exalead.com/voxalead/ and a video that describe the interface is available at: http://www.youtube.com/watch?v=1RhWGu9ElUUThis demonstrator has an optimized version for tablet (Ipad).

International audiencePhoto sharing platforms users often annotate their trip photos with landmark names. These annotations can be aggregated in order to recommend lists of popular visitor attractions similar to those found in classical tourist guides. However, individual tourist preferences can vary significantly so good recommendations should be tailored to individual tastes. Here we pose this visit personalization as a collaborative filtering problem. We mine the record of visited landmarks exposed in online user data to build a user-user similarity matrix. When a user wants to visit a new destination, a list of potentially interesting visitor attractions is produced based on the experience of like-minded users who already visited that destination. We compare our recommender to a baseline which simulates classical tourist guides on a large sample of Flickr users