This repo contains data on extended named entites (XNE) discussed in the 3rd chapter of the thesis of Noémi Ligeti-Nagy (The Right Edge of the Hungarian NP -- A Computational Approach. Pázmány Péter Catholic University, Doctoral School of Linguistics. Submitted in 2021)
The strings in the files are all retrieved from Szeged Treebank 2.0 (Csendes, Dóra; Csirik, János; Gyimóthy, Tibor; Kocsor, András 2005: The Szeged Treebank. In: Matoušek, Václav et al. (eds.): Proceedings of the 8th International Conference on Text, Speech and Dialogue (TSD 2005), Karlovy Vary, Czech Republic, September 12-16, 2005, Springer LNAI 3658, pp. 123-131.).
all_NOUN+NOUNs: list of all word combninations retrieved from the corpus. These are noun + noun pairs, where both are tagged as the members of the same noun phrase.
all_names: list of all extended named entities retrieved from the list "all_NOUN+NOUNs" by a manual filtering
XNEs_Szeged_lemmas_freq: The complete list of the lemmas of the names (the tokens St. Antonio herceg-nek 'St. Antonio prince-Dat' and St. Antonio herceg 'St. Antonio prince' are one type) in a frequency order. The list contains 902 types.
XNEs modified: list of extended named entities that consist of one or more proper nouns and a common noun and its modifier.
endings_lemma_sorted: List of the second parts of XNEs, the common noun, sorted by frequency.
w2v_try: folder containing the results of my attempt to collect as many members to the different groups of XNE endings as possible. I retrieved the 100 closest words (according to the word embedding vectors of these words (see Siklósi, B., and Novák, A. (2016a). Beágyazási modellek alkalmazása lexikai kategorizációsfeladatokra [Using word embedding models for lexical categorization]. In A. Tanács,V. Varga, and V. Vincze (Eds.) XII. Magyar Számítógépes Nyelvészeti Konferencia (MSZNY 2016), (pp. 3–14).) to the members of the different categories (néven, geographical common nouns, courtesy formulas, occupations, institution names and brand names and products). Then I used a clustering and visualisation tool (Novák, A., Siklósi, B., and Wenszky, N. (2017). Szóbeágyazási modellek vizualizációjára és böngészésére szolgáló webes felület [Online interface for the visualization and browsingof word embedding models]. In XIII. Magyar Számítógépes Nyelvészeti Konferencia, (pp. 355–362).) to group the results.