ALT at SemEval-2020 Task 12: Arabic and English Offensive Language Identification in Social Media

Proceedings of the 14th International Workshop on Semantic Evaluation, pages 201–207 Barcelona, Spain (Online), December 12, 2020. 201 EmbLexChange at SemEval-2020 Task 1: Unsupervised Embedding-based Detection of Lexical Semantic Changes Ehsaneddin Asgari†, Christoph Ringlstetter†, Hinrich Sch¨utze⋄ † NLP Expert Center, Data:Lab, Volkswagen AG, Munich, Germany ⋄ Center for Information and Language Processing, LMU Munich, Germany {ehsaneddin.asgari, christoph.ringlstetter}@volkswagen.de inquiries@cislmu.org Abstract This paper describes EmbLexChange, a system introduced by the “Life-Language” team for SemEval-2020 Task 1, on unsupervised detection of lexical-semantic changes. EmbLexChange is defined as the divergence between the embedding based profiles of word w (calculated with re- spect to a set of reference words) in the source and the target domains (source and target domains can be simply two time frames t1 and t2). The underlying assumption is that the lexical-semantic change of word w would affect its co-occurring words and subsequently alters the neighbor- hoods in the embedding spaces. We show that using a resampling framework for the selection of reference words (with conserved senses), we can more reliably detect lexical-semantic changes in English, German, Swedish, and Latin. EmbLexChange achieved second place in the binary detection of semantic changes in the SemEval-2020. 1 Introduction SemEval 2020 Task 1 is defined on the unsupervised detection of word sense changes over time in Ger- man, English, Swedish, and Latin. In particular, this challenge focused on detection and quantification of the sense changes of word w in the transition from the time period t1 to the time period t2 in the above mentioned four languages, where the input for each language are the text corpora dating from t1 and t2. This challenge involved two subtasks: i. Classification: The goal of the classification task is the binary detection of lexical semantic change implying loss or gain of senses (from t1 to t2) for the given word w. ii. Ranking: This subtask involves the ranking of lexical-semantic change for a given list of words (w1, w2, . . . , wM) by assigning scores quantifying relative changes of the word senses. In order to evaluate the two subtasks, the participating systems are evaluated against a ground truth corpus annotated by native speakers or scholars of the respective languages (Schlechtweg et al., 2020). Human languages constantly change due to cultural, technological, and social drift. Lexical semantic changes of human languages can materialize in the form of introducing/borrowing new words, or for the existing words can involve acquiring/losing some word senses (Koch, 2016; Traugott, 2017). Com- putational methods for automated detection of semantic changes can be extremely helpful in the study of historical texts or corpora spanning a very long period of time, e.g., semantic analysis of 1000 years of poetry (Asgari and Chappelier, 2013; Asgari et al., 2013), or in the design of the OCR algorithm for text digitization, or in designing an information retrieval system incorporating the semantic changes (Tahmasebi et al., 2018). Applications in the study of historical texts aside, the proposed methods detect lexical-semantic drift also in the same time period for …

Content truncated to 500 words