Using Twitter Streams for Opinion Mining: a case study on Airport Noise - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Communications in Computer and Information Science Année : 2020

Using Twitter Streams for Opinion Mining: a case study on Airport Noise

Résumé

This paper proposes a classification model for opinion mining around airport noise based on techniques such as event detection and sentiment analysis applied on Twitter posts. Tweets are retrieved using the Twitter API either because of location or content.A dataset of preprocessed, with NLP techniques, tweets is manually annotated and then used to train an SVM (Support Vector Machine) classifier in order to extract the relevant ones from the obtained collections. The extracted tweets from the SVM classifier are fed to a lexicon-based classifier to filter out the false relevant and to increase precision. A lexicon-based sentiment classifier is then applied in order to separate positive, negative and neutral tweets. The sentiment classifier uses emoticons, polarity of words with subjective intensity, intensifiers, negation effect with dynamic scope, contrast effect and SWN to detect the sentiment of tweets in a hierarchical manner. The information present in the classified tweets is used for a statistical survey-like study.
Fichier principal
Vignette du fichier
Meddeb_et_all-AirportNoise.pdf (524.89 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03018998 , version 1 (23-11-2020)

Identifiants

Citer

Iheb Meddeb, Catherine Lavandier, Dimitris Kotzinos. Using Twitter Streams for Opinion Mining: a case study on Airport Noise. Communications in Computer and Information Science, 2020, ⟨10.1007/978-3-030-44900-1_10⟩. ⟨hal-03018998⟩
264 Consultations
372 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More