Research Output
Articles
Publication Date: 2023
Onomazein (07171285)(61)pp. 191-211
In this article, we are going to introduce an automatic mechanism to intelligently extend the training set to improve the n-gram language model of Persian. Given the free word-order property in Persian, our enrichment algorithm diversifies n-gram combinations in baseline training data through dependency reordering, adding permissible sentences and filtering ungrammatical sentences using a hybrid empirical (heuristic) and linguistic approach. Experiments performed on baseline training set (taken from a standard Persian corpus) and the resulting enriched training set indicate a declining trend in average relative perplexity (between 34% to 73%) for informal/spoken vs. formal/written Persian test data. © 2023 Pontificia Universidad Catolica de Chile. All rights reserved.
Publication Date: 2023
Voprosy Jazykoznanija (0373658X)2023(5)pp. 133-147
In this article, we intend to examine the transferability of parts of English FrameNet to other lan-guages, specifically Persian. To accomplish this task, we consider the concept of motion due to its sig-nificance in cross-linguistic studies and its different lexicalization patterns in typologically diverse lan-guages. In this regard, we investigate the challenges that different features of the Persian language and particularly the difference in the way of expressing motion events in this language make for the compat-ibility of English FrameNet with Persian FrameNet. Focusing on Persian motion verbs, we try to deter-mine the extent to which it is possible to transfer different components of English FrameNet — including semantic frames, coreness and peripherality status of frame elements, syntactic and semantic structures and frame-to-frame relations — to other languages. The results suggest that English FrameNet is prone to incompatibility due to the lack of systematicity in the specification of frames and frame-to-frame re-lations. Moreover, different polysemy patterns, the presence of the same word as different lexical units, various syntactic alternations of lexical units, different levels of importance of frame elements in lexi-calization in various languages, the tendency of some languages like Persian to form compounds as well as the impact of culture make challenges for the transferability of English FrameNet to other languages. © 2023, Russian Academy of Sciences. All rights reserved.
Publication Date: 2021
Language Related Research (23223081)12(4)pp. 165-198
This article provides and examines empirical data from different languages showing that long A-movement of the subject of the embedded finite clause to the subject position of the main clause is indeed possible in many languages of the world. However all kinds of raising out of finte clause are not the same and have obvious differences from each other.In present article we introduce and acoount four kinds of finite raising including hyper raising, super raising further raising and copy raising in different languages. Finally we try to determine the position of Persian in cross-linguistic analysis of raising. © 2021, Tarbiat Modares University. All rights reserved.
Publication Date: 2020
Kervan (1825263X)24(1)pp. 107-122
This study aims to comparatively investigate the semantic frames of motion verbs in Persian and English within the framework of the frame semantics theory (Fillmore 1977; 1982; 1985). As far as motion verbs are concerned, Manner is considered as one of the motion components expressed by either the verb or any element other than the verb. In English — a satellite-framed language (Talmy 2000b) — Manner is shown by motion verbs, whereas in Persian it is typically indicated by non-verbal elements, although there are also some verbs via which Manner is encoded. Within this study, thirty English verbs of manner were selected from among the ones Levin (1993) has introduced and then the verbs were translated into Persian and looked up through the Persian Corpus of Bijankhan to achieve their contexts of use. Next, FrameNet was asked for the semantic frame each verb evoked. Thereafter, comparing the semantic frames in the two languages, it was revealed that not every verb of manner does exist as a Lexical Unit in FrameNet. Likewise, not for every verb was a specified semantic frame either. Moreover, the frames for some other verbs have been defined in such a way that they cannot semantically distinguish those verbs from each other, whereas such distinctions are prominent in both manner verbs and the frames they evoke especially in Persian. © 2020, Universita degli Studi di Torino, Facolta di Lingue e Letterature Straniere. All rights reserved.
Publication Date: 2019
Voprosy Jazykoznanija (0373658X)2019(3)pp. 101-124
This paper is an attempt to evaluate "manner of motion salience" in Ilami, a southern dialect of Kurdish. Manner components can cross-linguistically be encoded by main verbs, manner verbs, manner preverbs, and ideophones, among other linguistic devices. First, we will discuss the ways this semantic component is expressed in Ilami Kurdish. Then, in order to assess the salience of the manner component, a dictionary-based lexical survey and several experimental trials will be carried out. Following a study by F. E. Cardini ("Manner of motion saliency: An inquiry into Italian", 2008), we will conduct a vocabulary investigation to calculate the number of manner verbs in Ilami Kurdish, and then compare the statistics with Italian (a verb-framed language) and English (a satellite-framed language) to measure manner salience in Ilami Kurdish in comparison with other languages. We will also perform several experimental trials, aimed at evaluating manner salience in the actual use of this dialect. Accordingly, the experiments will be carried out to assess "ease of lexical access", which helps us to understand how readily participants (children and adults) can retrieve manner verbs from memory. Cross-linguistic analysis of Ilami Kurdish shows that this dialect is a relatively highly manner salient language, in which the elaboration of the manner component is rich and pervasive. This finding is supported by the dictionary-based lexical survey as well as ease of lexical access trials. Use of various strategies by Ilami Kurdish speakers to express manner sheds some light on the fact that manner salience is not solely affected by the lexicalization patterns of languages, as this component can also be encoded regardless of the path coding position. © 2019 Russian Academy of Sciences. All rights reserved.
