@inproceedings{lukes-sogaard-2018-sentiment,
    title = "Sentiment analysis under temporal shift",
    author = "Lukes, Jan  and
      S{\o}gaard, Anders",
    editor = "Balahur, Alexandra  and
      Mohammad, Saif M.  and
      Hoste, Veronique  and
      Klinger, Roman",
    booktitle = "Proceedings of the 9th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis",
    month = oct,
    year = "2018",
    address = "Brussels, Belgium",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/W18-6210/",
    doi = "10.18653/v1/W18-6210",
    pages = "65--71",
    abstract = "Sentiment analysis models often rely on training data that is several years old. In this paper, we show that lexical features change polarity over time, leading to degrading performance. This effect is particularly strong in sparse models relying only on highly predictive features. Using predictive feature selection, we are able to significantly improve the accuracy of such models over time."
}