In statistical machine translation, data sparsity is a challenging problem especially for languages with rich morphology and inconsistent orthography, such as Persian. We show that orthographic preprocessing and morphological segmentation of Persian verbs in particular improves the translation quality of Persian-English by 1.9 BLEU points on a blind test set.

In this paper, we describe an evaluation of the output of the translator using conceptbased grammars. This translator translates the Korean sentence generated by a speech recognizer into an English sentence through a concept analysis approach. A partial parsing function added to the translator and obtained better improvement because the performance of the parser (whole parser) is low in the sta...

Machine translation is the process of translating text from one natural language to other using computers. The process requires extreme intelligence and experience like a human being that a machine usually lacks. Availability of machine translators for translation from English to Dravidian language, Malayalam is on the low. A few corpus-based and non-corpus based approaches have been tried in p...

This study is an attempt to carry out a comparative analysis using Natural Semantic Metalanguage (henceforth NSM). The offering routine patterns of native Persian speakers was compared with that of Native American English speakers to see if it can provide evidence for applicability of NSM model which is claimed to be universal. The descriptive technique was the cultural scripts approach, using ...

In this paper, an automatic method for Persian WordNet construction based on Prenceton WordNet 2.1 (PWN) is introduced. The proposed approach uses Persian and English corpora as well as a bilingual dictionary in order to make a mapping between PWN synsets and Persian words. Our method calculates a score for each candidate synset of a given Persian word and for each of its translation, it select...

Bilingual parallel corpora are very important in various filed of natural language processing (NLP). The quality of a Statistical Machine Translation (SMT) system strongly dependent upon the amount of training data. For low resource language pairs such as Persian-English, there are not enough parallel sentences to build an accurate SMT system. This paper describes a new approach to use the Wiki...

The term metadiscourse rarely appears in translation studies despite the continuously growing body of research on discourse markers in different genres and through various perspectives. Translation as a product that needs to observe such markers for their communicative power and contribution to the overall coherence of a text within a context has not been satisfactorily studied. Motivated by su...

Parallel data are an important part of a reliable Statistical Machine Translation (SMT) system. The more of these data are available, the better the quality of the SMT system. However, for some language pairs such as Persian-English, parallel sources of this kind are scarce. In this paper, a bidirectional method is proposed to extract parallel sentences from English and Persian document aligned...

The study of compliments has attracted the attention of many scholars (e.g., Goffman 1971; Lakoff 1973; Brown and Levinson 1978; Amouzadeh 2001; Golato 2002; Sharifian 2005) and has become a major issue in the area of interactional sociolinguistics. To date, many models of politeness have been put forward in the literature. In this study, Brown and Levinson’s (1978, 1987) politeness model was u...