Online news comes in rich multimedia form: video, audio, photos, in addition to traditional text. Recent advances in semantically-rich text processing, in speech-to-text processing, and in image processing allows us to develop new ways of presenting and enriching news stories. Here we present MEWS, a Multimedia nEWS platform, which enriches news browsing according to media (text, images, and video) and to automatically detected type of news (music, general news, politics).

People often try to find an image using a short query and images are usually indexed using short annotations. Matching the query vocabulary with the indexing vocabulary is a difficult problem when little text is available. Textual user generated content in Web 2.0 platforms contains a wealth of data that can help solve this problem. Here we describe how to use Wikipedia and Flickr content to improve this match. The initial query is launched in Flickr and we create a query model based on co-occurring terms. We also calculate nearby concepts using Wikipedia and use these to expand the query. The final results are obtained by ranking the results for the expanded query using the similarity between their annotation and the Flickr model. Evaluation of these expansion and ranking techniques, over the Image CLEF 2010 Wikipedia Collection containing 237,434 images and their multilingual textual annotations, shows that a consistent improvement compared to state of the art methods.

Most news organizations provide immediate access to topical news broadcasts through RSS streams or podcasts. Until recently, applications have not permitted a user to perform content based search within a longer spoken broadcast to find the segment that might interest them. Recent progress in both automatic speech recognition (ASR) and natural language processing (NLP) has produced robust tools that allow us to now provide users with quicker and more focused access to relevant segments of news broadcast videos. Our public online demonstrator of the Voxalead application currently indexes daily broadcast news content from 50 sources in English, French, Chinese, Arabic, Spanish, Dutch, Italian and Russian.

Semantics has many different definitions in science. In natural language processing, there has been much research over the past three decades involving extracting the semantics, the meaning, of natural texts. This has led to entity recognition (people, places, companies, prices, dates, and events), and more recently into sentiment analysis, exploring another level of meaning in a text. These techniques are now well understood and robust. Results of this research are beginning to appear in products and online sites, finding their way into practical applications. The stage is set for an explosion of semantically savvy applications, from 3D design, to enhanced web browsing, to social network aware yellowpages. This talk will explore these paths from research to industry, illustrated by current products on the market.

Tourist photographs constitute a large part of the images uploaded to photo sharing platforms. But filtering methods are needed before one can extract useful knowledge from noisy user-supplied metadata. Here we show how to extract clean trip related information (what people visit, for how long, panoramic spots) from Flickr metadata. We illustrate our technique on a sample of metadata and images covering 183 cities of different size and from different parts of the world.

We present an interface to video and audio podcasts that extracts semantics from the speech content, and packages the extracted information in a variety of navigation tools. The user can jump to the relevant sections and browse from relevant section to relevant section. This interface is related to the Yahoo! Challenge: Robust Automatic Segmentation of Video According to Narrative Themes.

Geographical gazetteers are necessary in a wide variety of applications. In the past, the construction of such gazetteers has been a tedious, manual process and only recently have the first attempts to automate the gazetteers creation been made. Here we describe our approach for mining accurate but large-scale multilingual geographic information by successively filtering information found in heterogeneous data sources (Flickr, Wikipedia, Panoramio, Web pages indexed by search engines). Statistically cross-checking information found in each site, we are able to identify new geographic objects, and to indicate, for each one, its name, its GPS coordinates, its encompassing regions (city, region, country), the language of the name, its popularity, and the type of the object (church, bridge, etc.). We evaluate our approach by comparing, wherever possible, our multilingual gazetteer to other known attempts at automatically building a geographic database and to Geonames, a manually built gazetteer.

Uploading tourist photos is a popular activity on photo sharing platforms. These photographs and their associated metadata (tags, geo-tags, and temporal information) should be useful for mining information about the sites visited. However, user-supplied metadata are often noisy and efficient filtering methods are needed before extracting useful knowledge. We focus here on exploiting temporal information, associated with tourist sites that appear in Flickr. From automatically filtered sets of geo-tagged photos, we deduce answers to questions like "how long does it take to visit a tourist attraction?" or "what can I visit in one day in this city?" Our method is evaluated and validated by comparing the automatically obtained visit duration times to manual estimations.

At the NTCIR-4 workshop, Justsystem Corporation (JSC) and Clairvoyance Corporation (CC) collaborated in the cross-language retrieval task (CLIR). Our goal was to evaluate the performance and robustness of our recently developed commercial-grade CLIR systems for English and Asian languages. The main contribution of this article is the investigation of different strategies, their interactions in both monolingual and bilingual retrieval tasks, and their respective contributions to operational retrieval systems in the context of NTCIR-4. We report results of Japanese and English monolingual retrieval and results of Japanese-to-English bilingual retrieval. In monolingual retrieval analysis, we examine two special properties of the NTCIR experimental design (two levels of relevance and identical queries in multiple languages) and explore how they interact with strategies of our retrieval system, including pseudo-relevance feedback, multi-word term down-weighting, and term weight merging strategies. Our analysis shows that the choice of language (English or Japanese) does not have a significant impact on retrieval performance. Query expansion is slightly more effective with relaxed judgments than with rigid judgments. For better retrieval performance, weights of multi-word terms should be lowered. In the bilingual retrieval analysis, we aim to identify robust strategies that are effective when used alone and when used in combination with other strategies. We examine cross-lingual specific strategies such as translation disambiguation and translation structuring, as well as general strategies such as pseudo-relevance feedback and multi-word term down-weighting. For shorter title topics, pseudo-relevance feedback is a major performance enhancer, but translation structuring affects retrieval performance negatively when used alone or in combination with other strategies. All experimented strategies improve retrieval performance for the longer description topics, with pseudo-relevance feedback and translation structuring as the major contributors.

Multilingual applications frequently involve dealing with proper names, but names are often missing in bilingual lexicons.This problem is exacerbated for applications involving translation between Latin-scripted languages and Asian languages such as Chinese, Japanese and Korean (CJK) where simple string copying is not a solution.We present a novel approach for generating the ideographic representations of a CJK name written in a Latin script.The proposed approach involves first identifying the origin of the name, and then back-transliterating the name to all possible Chinese characters using language-specific mappings.To reduce the massive number of possibilities for computation, we apply a three-tier filtering process by filtering first through a set of attested bigrams, then through a set of attested terms, and lastly through the WWW for a final validation.We illustrate the approach with English-to-Japanese back-transliteration. Against test sets of Japanese given names and surnames, we have achieved average precisions of 73% and 90%, respectively.

The Web provides the largest, exploitable collection of language use. If we can mine the Web to build abstract models of language use, these models may have many applications. Here we present one example of using the implicit intelligence of language use to solve an important problem for machine translation programs and cross-lingual applications. This problem involves the translation of words written in katakana characters in Japanese. In this paper, we describe techniques of discovering katakana transliteration of English names and of finding English translations of multiword katakana sequences using implicit language models of English and Japanese found on the Web. These techniques were evaluated against human-constructed English-katakana glosses.

For cross language information retrieval (CLIR) based on bilingual translation dictionaries, good performance depends upon lexical coverage in the dictionary. This is especially true for languages possessing few inter-language cognates, such as between Japanese and English. In this paper, we describe a method for automatically creating and validating candidate Japanese transliterated terms of English words. A phonetic English dictionary and a set of probabilistic mapping rules are used for automatically generating transliteration candidates. A monolingual Japanese corpus is then used for automatically validating the transliterated terms. We evaluate the usage of the extracted English-Japanese transliteration pairs with Japanese to English retrieval experiments over the CLEF bilingual test collections. The use of our automatically derived extension to a bilingual translation dictionary improves average precision, both before and after pseudo-relevance feedback, with gains ranging from 2.5% to 64.8%.

The Web, teeming as it is with language data, of all manner of varieties and languages, in vast quantity and freely available, is a fabulous linguists' playground. This special issue of Computational Linguistics explores ways in which this dream is being explored.

No abstract available.

Article Free Access Share on Querying across languages: a dictionary-based approach to multilingual information retrieval Authors: David A. Hull Rank Xerox Research Centre, 6 chemin de Maupertuis, 38240 Meylan France Rank Xerox Research Centre, 6 chemin de Maupertuis, 38240 Meylan FranceView Profile , Gregory Grefenstette Rank Xerox Research Centre, 6 chemin de Maupertuis, 38240 Meylan France Rank Xerox Research Centre, 6 chemin de Maupertuis, 38240 Meylan FranceView Profile Authors Info & Claims SIGIR '96: Proceedings of the 19th annual international ACM SIGIR conference on Research and development in information retrievalAugust 1996 Pages 49–57https://doi.org/10.1145/243199.243212Published:18 August 1996Publication History 143citation1,652DownloadsMetricsTotal Citations143Total Downloads1,652Last 12 Months123Last 6 weeks9 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

Many of the processing steps in natural language engineering can be performed using finite state transducers. An optimal way to create such transducers is to compile them from regular expressions. This paper is an introduction to the regular expression calculus, extended with certain operators that have proved very useful in natural language applications ranging from tokenization to light parsing. The examples in the paper illustrate in concrete detail some of these applications.

For a very long time, it has been considered that the only way of automatically extracting similar groups of words from a text collection for which no semantic information exists is to use document co-occurrence data. But, with robust syntactic parsers that are becoming more frequently available, syntactically recognizable phenomena about word usage can be confidently noted in large collections of texts. We present here a new system called SEXTANT which uses these parsers and the finer-grained contexts they produce to judge word similarity.

One aspect of world knowledge essential to information retrieval is knowing when two words are related. Knowing word relatedness allows a system given a user's query terms to retrieve relevant documents not containing those exact terms. Two words can be said to be related if they appear in the same contexts Document co-occurrence gives a measure of word relatedness that has proved to be too rough to be useful. The relatively recent apparition of on-line dictionaries and robust and rapid parsers permits the extraction of finer word contexts from large corpora. In this paper, we will describe such an extraction technique that uses only coarse syntactic analysis and no domain knowledge. This technique produces lists of words related to any work appearing in a corpus. When the closest related terms were used in query expansion of a standard information retrieval testbed, the results were much better than that given by document co-occurence techniques, and slightly better than using unexpanded queries, supporting the contention that semantically similar words were indeed extracted by this technique.