This paper describes the workflow and architecture adopted by a linguistic research project. We report our experience and present the research outputs turned into resources that we wish to share with the community. We discuss the current limitations and the next steps that could be taken for the scaling and development of our research project. Allying NLP and language-centric AI, we discuss similar projects and possible ways to start collaborating towards potential platform interoperability.

Cet article décrit un prototype axé sur la prédiction du niveau de compétence des apprenants de l’anglais. Le système repose sur un modèle d’apprentissage supervisé, couplé à une interface web.

FinSentiA: Sentiment Analysis in English Financial Microblogs The objective of this paper is to report on the building of a Sentiment Analysis (SA) system dedicated to financial microblogs in English. The purpose of our work is to build a financial classifier that predicts the sentiment of stock investors in microblog platforms such as StockTwits and Twitter. Our contribution shows that it is possible to conduct such tasks in order to provide fine grained SA of financial microblogs. We extracted financial entities with relevant contexts and assigned scores on a continuous scale by adopting a deep learning method for the classification.

This paper focuses on aspect extraction which is a sub-task of Aspect-based Sentiment Analysis. The goal is to report an extraction method of financial aspects in microblog messages. Our approach uses a stock-investment taxonomy for the identification of explicit and implicit aspects. We compare supervised and unsupervised methods to assign predefined categories at message level. Results on 7 aspect classes show 0.71 accuracy, while the 32 class classification gives 0.82 accuracy for messages containing explicit aspects and 0.35 for implicit aspects.

This paper discusses the “Fine-Grained Sentiment Analysis on Financial Microblogs and News” task as part of SemEval-2017, specifically under the “Detecting sentiment, humour, and truth” theme. This task contains two tracks, where the first one concerns Microblog messages and the second one covers News Statements and Headlines. The main goal behind both tracks was to predict the sentiment score for each of the mentioned companies/stocks. The sentiment scores for each text instance adopted floating point values in the range of -1 (very negative/bearish) to 1 (very positive/bullish), with 0 designating neutral sentiment. This task attracted a total of 32 participants, with 25 participating in Track 1 and 29 in Track 2.

Improving lexical network’s quality is an important issue in the creation process of these language resources. This can be done by automatically inferring new relations from already existing ones with the purpose of (1) densifying the relations to cover the eventual lack of information and (2) detecting errors. In this paper, we devise such an approach applied to the JeuxDeMots lexical network, which is a freely available lexical and semantic resource for French. We first present the principles behind the lexical network construction with crowdsourcing and games with a purpose and illustrated them with JeuxDeMots (JDM). Then, we present the outline of an elicitation engine based on an inference engine using schemes like deduction, induction and abduction which will be referenced and briefly presented and we will especially highlight the new scheme (Relation Inference Scheme with Refinements) added to our system. An experiment showing the relevance of this scheme is then presented.