Search

A finalised digital resource of 88,000 anonymised French text messages, the 88milSMS corpus, two extracts (1,000 SMS transcoded into standardised French and 100 linguistically annotated SMS) and sociolinguistic questionnaire data were released in June 2014 for all to download via a user free-of-charge licence agreement, from the Huma-Num web service (http://88milsms.huma-num.fr, Panckhurst et al., 2014). The sud4science project (http://sud4science.org, Panckhurst et al. 2013), enabling authentic text message collection from the general public by a group of academics, is part of a vast international initiative (http://www.sms4science.org/, Fairon et al. 2006, Cougnon and Fairon, 2014, Cougnon 2015), to build a worldwide database and analyse authentic text messages in different languages. We decided to exclude full transcoding and annotation tagging in the final corpus. This is a theoretical position, since annotation is far from neutral, and is invariably linked to an interpretative framework. Owing to varying theoretical disciplinary and scientific stances, it seems that a true consensus on how to standardise the transcoding and linguistic annotation tagging does not exist (Panckhurst, 2015). Other researchers may disagree and prefer to provide both 'raw' and fully tagged corpora (Chanier et al. 2014). This theoretical position does not exclude exploring Natural Language Processing (NLP) investigation techniques, which could then be implemented in real-life applications. Examples of investigation techniques are indicated as follows: 1) Our corpus can be used to analyse current mediated electronic discourse, and help build knowledge on different SMS writing forms (Roche et al. 2015). 2) Algorithms may be used to learn from this: alignment methods for facilitating automatic transcoding have been explored (Aw et al. 2006, Beaufort et al., 2008, Guimier de Neef and Fessard, 2007, Kobus et al, 2008, Lopez et al, 2014). 3) We have devised a method for classifying 'unknown' items within text messages, which may help to automatically identify lexical 'creativity' within 88milSMS and improve electronic dictionary approaches (Lopez et al. 2015). In order to refine automatic normalisation techniques for initially non-standard texts in French, the next logical step is to compare our resource with different types of instant media (i.e. SMS, forums, tweets). Firstly, a new typology of the detected 'mistakes', based on existing typologies, will be elaborated. Secondly, automatic normalisation techniques — focusing on the most frequent errors — will be proposed. These will then be confronted with traditional automatic translation (Vilariño et al., 2012), speech recognition (Kobus et al., 2008) and spelling/grammatical checker principles (Beaufort et al., 2010). Finally, the approach should enable comparison between different types of instant media.

This article highlights an approach based on authentic data, by focusing on recent research related to collection, processing and analysis of a large French text-message corpus, entitled 88milSMS (http://88milsms.huma-num.fr/, Panckhurst, Détrie, Lopez, Moïse, Roche, Verine, 2014), including a sociolinguistic questionnaire submitted to donors (with their answers). The authors, using a pluridisciplinary approach (linguistics/ language sciences, computer science, Natural Language Processing), explain why they chose to give the scientific community and the general public access to the SMS corpus., Nous présentons notre approche fondée sur les données authentiques, en nous concentrant sur des recherches récentes, portant sur le recueil, le traitement et l’analyse d’un grand corpus de SMS en français, intitulé 88milSMS (http://88milsms. huma-num.fr/, Panckhurst, Détrie, Lopez, Moïse, Roche, Verine, 2014), incluant un questionnaire sociolinguistique soumis aux donateurs au moment de la collecte ainsi que leurs réponses. Puis nous expliquons pourquoi, dans une démarche pluridisciplinaire (située entre sciences du langage, informatique et traitement automatique du langage naturel), nous avons décidé de fournir à la communauté scientifique et au grand public le corpus de SMS., Panckhurst Rachel, Roche Mathieu, Lopez Cédric, Verine Bertrand, Détrie Catherine, Moïse Claudine. De la collecte à l’analyse d’un corpus de SMS authentiques : une démarche pluridisciplinaire. In: Histoire Épistémologie Langage, tome 38, fascicule 2, 2016. Constitution de corpus linguistiques et pérennisation des données. pp. 73-85.

International audience; The sud4science LR project (http://www.sud4science.org/) aimed at studying a fairly recent form of written communication: SMS (Short Message Service). The first step of the project was to collect a large number of text messages from the general public. We initially gathered 93'085 SMS and our final corpus, entitled 88milSMS, contains over 88'000 SMS.2 In this article, we propose a novel approach(which is also applicable to other textual data)for classifying unknown items in 88milSMS, based on two steps: 1) Classification of SMS in relation to 5 European languages (French, Spanish, English, German, Italian), 2) Classification of unknown items according to predefined classes (schedules, items containing special character(s), number(s), words without accents, or with repeated characters, etc.). We are then able to make a distinction between the truly "original" items which are widely used compared to those that are rarely used in the corpus. Based on examples mined in the different classes, we present a preliminary analysis of the obtained resource.

International audience; This paper presents two anonymisation methods to process an SMS corpus. The first one is based on an unsupervised approach called Seek&Hide. The implemented system uses several dictionaries and rules in order to predict if a SMS needs anonymisation process. The second method is based on a supervised approach using machine learning techniques. We evaluate the two approaches and we propose a way to use them together. Only when the two methods do not agree on their prediction, will the SMS be checked by a human expert. This greatly reduces the cost of anonymising the corpus.

September 2-6, 2013; International audience; In Human Computer Interaction, it is more and more clear that usability is not enough. In order to take into account the other criteria that may be relevant for design, G. Cockton introduced the notion of "worth" and the Worth Centered Design (WCD) framework for its operationalization. The WCD framework structures the development process and provides designers with a set of tools, including Worth Maps (WMs). Worth maps connect systems attributes to human ones, and as such represent a promising tool. However, they remain understudied and under-experimented. This paper presents the results of our experience with WMs. More precisely, it proposes the PEW (Perceived and Expected Worth) framework for worth mapping, reports findings from a study conducted with 5 experts regarding many aspects of WMs, and discusses future directions for research. Keywords: Interactive systems design, worth, Worth Maps (WMs).

International audience; Although many new methodological and modelling concepts have been proposed by the scientific community, current industries are still focusing their engineering design process on CAD model since they assume it is the starting point of many analyses with respect to product life cycle (CAM, FEA, LCA…). The paper presents the application of modelling concepts that lead the progressive justification of CAD model with respect to knowledge synthesis by least commitment. Design experts are first formalizing their knowledge that is therefore translated to form features and parameters (topology, position, orientation, dimensions…). The results show that this new design approach and models support design intents and rational, but the generated CAD model is not fully justified. That drives to many conclusions: CAD model is many often non-100% rational by designers’ knowledge, design solution space is therefore larger than the one modelled in CAD software and could be used to foster innovation.

International audience; There has been increasing interest in the artificial intelligence community for influencer detection in recent years for its utility in singling out pertinent users within a large network of social media users. This could be useful, for example in commercial campaigns, to promote a product or a brand to a relevant target set of users. This task is performed either by analysing the graph-based representation of user interactions in a social network or by measuring the impact of the linguistic content of user messages in online discussions. We performed independent studies for each of these methods in the present paper with a hybridisation perspective. In the first study, we extract structural information to highlight influence among interaction networks. In the second, we identify linguistic features of influential behaviours. We then compute a score of user influence using centrality measures with the structural information for the former and a machine learning approach based on the relevant linguistic features for the latter.

International audience; Dans une base de connaissance, les entités se veulent pérennes mais certains événements induisent que les relations entre ces entités sont instables. C'est notamment le cas pour des relations entre organisations, produits, ou marques, entités qui peuvent être rachetées. Dans cet article, nous proposons une approche permettant d'extraire des relations d'appartenance entre deux entités afin de peu-pler une base de connaissance. L'extraction des relations à partir d'une source dynamique d'informations telle que Twitter permet d'atteindre cet objectif en temps réel. L'approche consiste à modéliser les événements en s'appuyant sur une ressource lexico-sémantique. Une fois les entités liées au Web des données ouvertes (en particulier DBpedia), des règles linguistiques sont appliquées pour finalement générer les triplets RDF qui représentent les événements.

International audience; Hyper-parameter tuning is a resource-intensive task when optimizing classification models. The commonly used k-fold cross validation can become intractable in large scale settings when a classifier has to learn billions of parameters. At the same time, in real-world, one often encounters multi-class classification scenarios with only a few labeled examples; model selection approaches often offer little improvement in such cases and the default values of learners are used. We propose bounds for classification on accuracy and macro measures (precision, recall, F1) that motivate efficient schemes for model selection and can benefit from the existence of unlabeled data. We demonstrate the advantages of those schemes by comparing them with k-fold cross validation and hold-out estimation in the setting of large scale classification.

National audience; Qu’est-ce que la donnée écrite en sciences du langage ? Trois types se distinguent : 1) la donnée lexicale, qui se présente essentiellement sous forme d’une entrée lexicale, regroupant un ensemble de propriétés ; 2) » le nom spécifique de la donnée observable en linguistique est l’exemple » et renvoie à « un énoncé qui pourrait être effectivement prononcé, même s’il ne l’est pas dans les faits » (Milner 1989, p. 51-52) ; 3) la donnée en tant que texte brut, i.e. le corpus. En linguistique(s) de corpus, il s’agit d’analyser les productions authentiques contenues dans le corpus. Dans certaines écoles linguistiques, au contraire, l’étude du corpus tout-venant n’a pas lieu d’être. Ainsi, perdure le débat concernant l’opposition (ou, tout au moins, la différenciation) entre exemples linguistiques (éventuellement « fabriqués ») et productions authentiques relevées dans des corpus (cf. entre autres, pour le français, Bilger et al. 2000, Cori et al. 2008, Habert et al.1997, Péry-Woodley 1995). En vingt ans, notre propre approche a évolué : d’une analyse linguistique-informatique basée sur l’exemple (Panckhurst 1994, p. 39), nous sommes passée à une analyse de la donnée authentique figurant dans des corpus (Panckhurst 2013, p. 97, Panckhurst et al. 2014). Pour nous, cette mutation s’explique, d’une part, par l’évolution de l’accès aux données, et, d’autre part, par le discours électronique médié (Panckhurst 1997, 2006), circulant entre individus se servant d’outils électroniques (ordinateurs, tablettes, téléphones portables, etc.), qui induit des pratiques et des usages émergents. En deux décennies, la constitution de corpus numérisés ou nativement numériques est devenue monnaie courante, et cette accessibilité massive constitue en soi une nouveauté. Les données authentiques existant sous la forme de courriels, forums, chats, blogs, réseaux sociaux, et, plus récemment de SMS, facilement exploitables par les chercheurs, permettent l’observation, la fouille et l’analyse des pratiques et des usages (novateurs ou non) des scripteurs. Dans le cadre de cette communication, nous expliquerons ce cheminement, en nous focalisant sur des recherches récentes, portant sur le recueil, le traitement et l’analyse d’un grand corpus de SMS en français, intitulé « 88milSMS » (consultable sur la grille de services d’Huma-Num : http://88milsms.huma-num.fr/).

International audience; In this paper, we propose a method for aligning text messages (entitled AlignSMS) in order to automatically build an SMS dictionary. An extract of 100 text messages from the 88milSMS corpus (Panckhurst el al., 2013, 2014) was used as an initial test. More than 90,000 authentic text messages in French were collected from the general public by a group of academics in the south of France in the context of the sud4science project (http://www.sud4science.org). This project is itself part of a vast international SMS data collection project, entitled sms4science (http://www.sms4science.org, Fairon et al. 2006, Cougnon, 2014). After corpus collation, pre-processing and anonymisation (Accorsi et al., 2012, Patel et al., 2013), we discuss how "raw" anonymised text messages can be transcoded into normalised text messages, using a statistical alignment method. The future objective is to set up a hybrid (symbolic/statistic) approach based on both grammar rules and our statistical AlignSMS method.

International audience; A great deal of research on opinion mining and sentiment analysis has been done in specific contexts such as movie reviews, commercial evaluations, campaign speeches, etc. In this paper, we raise the issue of how appropriate these methods are for documents related to land-use planning. After highlighting limitations of existing proposals and discussing issues related to textual data, we present the method called Opiland (OPinion mIning from LAND-use planning documents) designed to semi-automatically mine opinions in specialized contexts. Experiments are conducted on a land-use planning dataset, and on three datasets related to others areas highlighting the relevance of our proposal.

In the literature, automatic summarization and automatic titling tasks are often merged. It seems that a short summary can be considered as a relevant title. But can we compare a title and a summary without having previously studied their criteria? This study aims at positioning the emergent task of automatic titling with regard to automatic summarization task. In this paper, we define a set of criteria according to the summary and to the title, and we analyze the results obtained with our method of automatic classification. This analysis enables to report real objectives of both tasks and to validate their relevance. Dans la littérature, les tâches de résumé et de titrage automatique sont souvent confondues. A priori, il semble qu’un résumé de quelques mots peut constituer un titre tout à fait pertinent. Mais peut-on comparer un titre et un résumé sans auparavant en avoir étudié leurs critères ? Cette étude a pour but de positionner l’émergente tâche de titrage automatique par rapport à celle de résumé automatique. Après avoir défini les critères attachés au résumé et au titre, nous analysons les résultats obtenus via notre méthode automatique de classification, permettant de rendre compte des objectifs réels des deux tâches et de valider leur pertinence.

International audience; Automatic titling of text documents is an essential task for several applications (automatic heading of e-mails, summarization, and so forth). This paper describes a system facilitating information retrieval in a set of textual documents by tackling the automatic titling and subtitling issue. Automatic titling here involves providing both informative and catchy titles. We thus propose two different approaches based on NLP, text mining, and Web Mining techniques. The first one (POSTIT) consists of extracting relevant noun phrases from texts as candidate titles. An original approach combining statistical criteria and noun phrase positions in the text helps in collecting informative titles and subtitles. The second approach (NOMIT) is based on various assumptions made on POSTIT and aims to generate both informative and catchy titles. Both approaches are applied to a corpus of news articles, then evaluated according to two criteria, i.e. informativeness and catchiness.

This article describes the sud4science project (www.sud4science.org). Firstly, the authors present the acquisition phase of both SMS data and questionnaire data. Secondly, they explain anonymisation techniques, transcoding and optional annotation phases. Finally, they propose preliminary (socio-) linguistic analyses of scriptural usage of SMS writing, and they also indicate those that are planned in the foreseeable future., Dans le cadre de cet article, on expose le déroulement du projet sud4science (www.sud4science.org). En premier lieu, on décrit la phase d'acquisition des données en provenance des SMS et du questionnaire, avant d'aborder les étapes successives d'anonymisation, de transcodage et d'annotation optionnelle. Ensuite, on présente les analyses (socio-)linguistiques des pratiques scripturales de l'écriture SMS (eSMS) qui ont débuté, ainsi que celles prévues à court et à moyen terme.