My co-authored paper on temporal orientation of texts appeared in Research & Politics. In this study we applied latent semantic scaling (LSS) to a corpus of English and German texts to identify features related to the future or the past automatically. Only with a set common verbs as seed words, the algorithm could classify sentences from manifestos 60% to 70% accurately. Replication code is published in the Dataverse.
We also applied the same methodology to Twitter posts by German political parties to reveal policy issues’ temporal orientation. The result showed that issues such as Education, Economy and Environment are strongly about future, while Domestic Security and Foreign Policies are mainly about the past or present. We found that the result is intuitively correct.
I hope that our cost-efficient method makes temporal analysis of text more accessible and add an interesting dimension to political text analysis.
