In recent years there has been a growing interest in research about opinions, attitudes, sentiment and emotions expressed in written texts, involving for example on-line editions of print-based newspapers and newer media such as blogs or e-forums. A new field of research is rapidly establishing itself in a number of areas, pursuing a range of objectives and including the study of opinions about a variety of entities (products, behaviours and so on) based on methods and techniques that are equally diverse. Evidence for this increase in interest includes the establishment of a number of research groups and the recent opinion mining exercises FODDP ’08, DEFT ’07, DEFT ’09, CTNIR6. The study of opinions, attitudes and sentiment involves a diverse range of variables: the object of opinion, the source of opinion (made complex by the presence of attribution and dialogue), the strength and polarity of the opinion, whether it is expressed overtly or covertly, the degree of alignment in opinion between the text producer and the receiver. It is known that the expression of opinion is complex and can be fine-grained. As a result, attempts to focus only on the simple assignment of polarity to a whole document lead to results which are exploitable only in relatively restricted cases. Alongside this variation goes a variety of research approaches, including the automation of sentiment identification and analysis, the investigation of stance in conversation and in written text, the modelling of evaluative language in Appraisal theory (J. Martin and P. White 2005 The Language of Evaluation), and the use of corpus-driven lexical studies to identify evaluative prosodies. To these might be added studies of evidentiality, metadiscourse, point of view, rhetoric, persuasion and argumentation, and philosophical approaches to the activity of evaluation. There is an increasing amount of data available to researchers. For example, on many websites, visitors are invited to write or say what they think of a product, an object, an idea or a person. There are no constraints on the kind of language that can be used, and the extreme variability of forms which express subjective points of view is such that they are particularly difficult to detect automatically. There is no agreed approach that would process this data and make it of benefit to the user. In view of the wide variety of work that has been undertaken from a range of disciplines, and the need to achieve more applicable results, a review and proposals for future work is timely. It needs to be undertaken from at least four points of view: (1) a modelling of how knowledge comes about (what is an ‘opinion’, how can it be represented?); (2) expression in language and discourse (how are opinions, in their various guises, formulated?); (3) the methods needed to achieve an automatic analysis of written and recorded texts and (4) how a diversity of opinions can be identified and represented.
We wish to publish both innovative papers and those which synthesise previous work and look to the future on themes including the following (indicative list only):
Modelling opinions using linguistics and information science;
The construction, acquisition and validation of linguistics resources (lexical, grammatical, phraseological) for the expression of opinions;
The characterisation, annotation and automatic extraction of opinions in written and audiovisual texts;
The representation of opinion-bearing discourse;
The study and treatment of opinions in commercial contexts.
TAL (Traitement Automatique des Langues / Natural Language Processing) is a forty year old international journal published by ATALA (French Association for Natural Language Processing) with the support of CNRS (National Centre for Scientific Research). It has moved to an electronic mode of publication, with printing on demand. This affects in no way its reviewing and selection process.
Agata Jackiewicz (STIH, Université de Paris-Sorbonne) Susan Hunston (University of Birmingham) Marc El-Beze (LIA, Université d’Avignon)
Authors intending to submit a paper are encouraged to contact the chief editors of this special issue.
Contributions (25 pages maximum, PDF format) must be sent by e-mail to the addresses below:
Style sheets are available for download on the Web site of the journal.
31/5/2010 Submission deadline
15/9/2010 Decision of the editorial board
1/11/2010 Revised version of the accepted papers
Publication is expected by the beginning of 2011
Nicholas Asher, IRIT, Université Paul Sabatier, Toulouse
Monika Bednarek, Université de Sidney
Farah Benamara Zitoune, IRIT, Université Paul Sabatier, Toulouse
Yves Bestgen, CECL, Université catholique de Louvain
Thierry Charnois, GREYC, Université de Caen
Chloé Clavel, EDF, Paris
Stéphane Ferrari, GREYC, Université de Caen
Catherine Gouttas, Thales
Philippe Laublet, STIH, Université Paris-Sorbonne
Mark Lee, Université de Birmingham
Dominique Legallois, CRISCO, Université de Caen
Marie-Francine Moens, Katholieke Universiteit Leuven
Laura Monceaux, LINA, Université de Nantes
Renato De Mori, LIA, Université d’Avignon
Maria Teresa Pazienza, Université de Rome Tor Vergata
Mathieu Roche, LIRMM, Université de Montpellier 2
Horacio Rodriguez , Université Polytechnique de Catalogne, Barcelone
Horacio Saggion, Université de Sheffield
Geoff Thompson, Université de Liverpool