Remarks on Compositionality (with reference to Gennadij Zeldoviè's article “On Russian Dative Reflexive Constructions: Accidental or Compositional”)

Zeldovič’s article “On Russian Dative Reflexive Constructions: Accidental or Compositional” is very interesting. It contains a good deal of insightful observations and is painstakingly argued. Its research object is the Russian dative reflexive construction (DRC) like Ивану не работается ‘Ivan does not feel like reading’. The aim of the article is to show that the DRC is fully compositional. Like many other works by Zeldovič, the article is written from the radical-pragmatic perspective and constitutes a very good illustration of this trend in linguistic research. The language material that it analyzes has often been investigated within more traditional frameworks, especially in Russian linguistics, which makes Zeldovič’s novel approach to the old problem particularly interesting. In this short note I would like (by way of discussion) to address two problems connected not so much with the DRC itself as with methodological issues concerning compositionality. I will dwell on two aspects: on the question of how we understand the very concept of compositionality, and what instruments we employ to demonstrate it.

Language identification based on a discriminative text categorization technique

In this paper, we describe new results and improvements to a lan-guage identification (LID) system based on PPRLM previously introduced in [1] and [2]. In this case, we use as parallel phone recognizers the ones provided by the Brno University of Technology for Czech, Hungarian, and Russian lan-guages, and instead of using traditional n-gram language models we use a lan-guage model that is created using a ranking with the most frequent and discrim-inative n-grams. In this language model approach, the distance between the ranking for the input sentence and the ranking for each language is computed, based on the difference in relative positions for each n-gram. This approach is able to model reliably longer span information than in traditional language models obtaining more reliable estimations. We also describe the modifications that we have being introducing along the time to the original ranking technique, e.g., different discriminative formulas to establish the ranking, variations of the template size, the suppression of repeated consecutive phones, and a new clus-tering technique for the ranking scores. Results show that this technique pro-vides a 12.9% relative improvement over PPRLM. Finally, we also describe re-sults where the traditional PPRLM and our ranking technique are combined.