Cross-lingual Structure Transfer for Zero-resource Event Extraction

Cross-lingual Structure Transfer for Zero-resource Event Extraction. Lu, D., Subburathinam, A., Ji, H., May, J., Chang, S., Sil, A., & Voss, C. In Proceedings of the 12th Language Resources and Evaluation Conference, pages 1976–1981, Marseille, France, May, 2020. European Language Resources Association.

Paper abstract bibtex

Most of the current cross-lingual transfer learning methods for Information Extraction (IE) have been only applied to name tagging. To tackle more complex tasks such as event extraction we need to transfer graph structures (event trigger linked to multiple arguments with various roles) across languages. We develop a novel share-and-transfer framework to reach this goal with three steps: (1) Convert each sentence in any language to language-universal graph structures; in this paper we explore two approaches based on universal dependency parses and complete graphs, respectively. (2) Represent each node in the graph structure with a cross-lingual word embedding so that all sentences in multiple languages can be represented with one shared semantic space. (3) Using this common semantic space, train event extractors from English training data and apply them to languages that do not have any event annotations. Experimental results on three languages (Spanish, Russian and Ukrainian) without any annotations show this framework achieves comparable performance to a state-of-the-art supervised model trained from more than 1,500 manually annotated event mentions.

@inproceedings{lu-etal-2020-cross,
    title = "Cross-lingual Structure Transfer for Zero-resource Event Extraction",
    author = "Lu, Di  and
      Subburathinam, Ananya  and
      Ji, Heng  and
      May, Jonathan  and
      Chang, Shih-Fu  and
      Sil, Avi  and
      Voss, Clare",
    booktitle = "Proceedings of the 12th Language Resources and Evaluation Conference",
    month = may,
    year = "2020",
    address = "Marseille, France",
    publisher = "European Language Resources Association",
    url = "https://www.aclweb.org/anthology/2020.lrec-1.243",
    pages = "1976--1981",
    abstract = "Most of the current cross-lingual transfer learning methods for Information Extraction (IE) have been only applied to name tagging. To tackle more complex tasks such as event extraction we need to transfer graph structures (event trigger linked to multiple arguments with various roles) across languages. We develop a novel share-and-transfer framework to reach this goal with three steps: (1) Convert each sentence in any language to language-universal graph structures; in this paper we explore two approaches based on universal dependency parses and complete graphs, respectively. (2) Represent each node in the graph structure with a cross-lingual word embedding so that all sentences in multiple languages can be represented with one shared semantic space. (3) Using this common semantic space, train event extractors from English training data and apply them to languages that do not have any event annotations. Experimental results on three languages (Spanish, Russian and Ukrainian) without any annotations show this framework achieves comparable performance to a state-of-the-art supervised model trained from more than 1,500 manually annotated event mentions.",
    language = "English",
    ISBN = "979-10-95546-34-4",
}

Downloads: 0

{"_id":"kAzdPe37etg44Lky2","bibbaseid":"lu-subburathinam-ji-may-chang-sil-voss-crosslingualstructuretransferforzeroresourceeventextraction-2020","author_short":["Lu, D.","Subburathinam, A.","Ji, H.","May, J.","Chang, S.","Sil, A.","Voss, C."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Cross-lingual Structure Transfer for Zero-resource Event Extraction","author":[{"propositions":[],"lastnames":["Lu"],"firstnames":["Di"],"suffixes":[]},{"propositions":[],"lastnames":["Subburathinam"],"firstnames":["Ananya"],"suffixes":[]},{"propositions":[],"lastnames":["Ji"],"firstnames":["Heng"],"suffixes":[]},{"propositions":[],"lastnames":["May"],"firstnames":["Jonathan"],"suffixes":[]},{"propositions":[],"lastnames":["Chang"],"firstnames":["Shih-Fu"],"suffixes":[]},{"propositions":[],"lastnames":["Sil"],"firstnames":["Avi"],"suffixes":[]},{"propositions":[],"lastnames":["Voss"],"firstnames":["Clare"],"suffixes":[]}],"booktitle":"Proceedings of the 12th Language Resources and Evaluation Conference","month":"May","year":"2020","address":"Marseille, France","publisher":"European Language Resources Association","url":"https://www.aclweb.org/anthology/2020.lrec-1.243","pages":"1976–1981","abstract":"Most of the current cross-lingual transfer learning methods for Information Extraction (IE) have been only applied to name tagging. To tackle more complex tasks such as event extraction we need to transfer graph structures (event trigger linked to multiple arguments with various roles) across languages. We develop a novel share-and-transfer framework to reach this goal with three steps: (1) Convert each sentence in any language to language-universal graph structures; in this paper we explore two approaches based on universal dependency parses and complete graphs, respectively. (2) Represent each node in the graph structure with a cross-lingual word embedding so that all sentences in multiple languages can be represented with one shared semantic space. (3) Using this common semantic space, train event extractors from English training data and apply them to languages that do not have any event annotations. Experimental results on three languages (Spanish, Russian and Ukrainian) without any annotations show this framework achieves comparable performance to a state-of-the-art supervised model trained from more than 1,500 manually annotated event mentions.","language":"English","isbn":"979-10-95546-34-4","bibtex":"@inproceedings{lu-etal-2020-cross,\n title = \"Cross-lingual Structure Transfer for Zero-resource Event Extraction\",\n author = \"Lu, Di and\n Subburathinam, Ananya and\n Ji, Heng and\n May, Jonathan and\n Chang, Shih-Fu and\n Sil, Avi and\n Voss, Clare\",\n booktitle = \"Proceedings of the 12th Language Resources and Evaluation Conference\",\n month = may,\n year = \"2020\",\n address = \"Marseille, France\",\n publisher = \"European Language Resources Association\",\n url = \"https://www.aclweb.org/anthology/2020.lrec-1.243\",\n pages = \"1976--1981\",\n abstract = \"Most of the current cross-lingual transfer learning methods for Information Extraction (IE) have been only applied to name tagging. To tackle more complex tasks such as event extraction we need to transfer graph structures (event trigger linked to multiple arguments with various roles) across languages. We develop a novel share-and-transfer framework to reach this goal with three steps: (1) Convert each sentence in any language to language-universal graph structures; in this paper we explore two approaches based on universal dependency parses and complete graphs, respectively. (2) Represent each node in the graph structure with a cross-lingual word embedding so that all sentences in multiple languages can be represented with one shared semantic space. (3) Using this common semantic space, train event extractors from English training data and apply them to languages that do not have any event annotations. Experimental results on three languages (Spanish, Russian and Ukrainian) without any annotations show this framework achieves comparable performance to a state-of-the-art supervised model trained from more than 1,500 manually annotated event mentions.\",\n language = \"English\",\n ISBN = \"979-10-95546-34-4\",\n}\n\n","author_short":["Lu, D.","Subburathinam, A.","Ji, H.","May, J.","Chang, S.","Sil, A.","Voss, C."],"key":"lu-etal-2020-cross","id":"lu-etal-2020-cross","bibbaseid":"lu-subburathinam-ji-may-chang-sil-voss-crosslingualstructuretransferforzeroresourceeventextraction-2020","role":"author","urls":{"Paper":"https://www.aclweb.org/anthology/2020.lrec-1.243"},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://jonmay.github.io/webpage/cutelabname/cutelabname.bib","dataSources":["ZdhKtP2cSp3Aki2ge","X5WBAKQabka5TW5z7","hbZSwot2msWk92m5B","fcWjcoAgajPvXWcp7","GvHfaAWP6AfN6oLQE","j3Qzx9HAAC6WtJDHS","5eM3sAccSEpjSDHHQ"],"keywords":[],"search_terms":["cross","lingual","structure","transfer","zero","resource","event","extraction","lu","subburathinam","ji","may","chang","sil","voss"],"title":"Cross-lingual Structure Transfer for Zero-resource Event Extraction","year":2020}