On the Importance of Ezafe Construction in Persian Parsing

Nourian, Alireza; Rasooli, Mohammad Sadegh; Imany, Mohsen; Faili, Heshaam

doi:10.3115/v1/p15-2144

Cited by 10 publications

(12 citation statements)

References 11 publications

(9 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tables and compare those algorithms. With respect to POS tagging, the effect of FNLP‐ontology is studied on the work of Nourian et al (). The gold Ezafe tag set introduced by the authors is extremely helpful in part of semantic analysis.…”

Section: Resultsmentioning

confidence: 99%

“…Statistical methods annotate words, employing the tags of their neighbours and the frequency of sequences. Conditional random fields (CRFs) are also used to annotate word sequences (McCallum & Lafferty, ; Nourian, Rasooli, Imany, & Faili, ). Research has shown that the CRFs generated more fault than statistical methods in early runs, although they produced more accurate outcomes after that.…”

Section: Related Workmentioning

confidence: 99%

“…Nourian et al () successfully used dependency grammar in Ezafe detection and reported a considerable increase in its precision rate. A dependency parser was also built for improving NLP tasks (Seraji, Megyesi, & Nivre, ).…”

Section: Related Workmentioning

confidence: 99%

“…An example could be on Ezafe detection. Ezafe is connected to a head noun and is followed by modifiers such as adjectival phrases and attributive nouns (Nourian et al, ). Some other relations can be represented using hierarchy diagrams (Galton, ).…”

Section: Language Rules and Ontology Designingmentioning

confidence: 99%

See 3 more Smart Citations

FNLP‐ONT: A feasible ontology for improving NLP tasks in Persian

2018

View full text Add to dashboard Cite

Natural language processing is a composition of several error‐prone and challenging tasks, including part of speech tagging, word sense disambiguation, named entity recognition, and compound verb detection. Studying intrasentence relations and roles is essential to improve the mentioned subtasks. Semi‐automatic schemes such as ontologies can be applied to clarify word's dependencies. This paper presents an ontology that is targeting to improve POS tagging, WSD, NER, and compound verb detection in Persian with extra properties that may ameliorate machine translation. The ontology is tested in combinations with several state‐of‐art algorithms on Dadegan corpus. The results show that coping semantic analysis with machine learning methods enhance relation detection and consequently precision of the mentioned subtasks, which is not widely addressed in Persian. Furthermore, the experimental results declare that the accuracy rate increases between 4.5 and 23% for different tasks.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Language Rules and Ontology Designingmentioning

confidence: 99%

See 2 more Smart Citations

FNLP‐ONT: A feasible ontology for improving NLP tasks in Persian

2018

View full text Add to dashboard Cite

show abstract

“…Compare Examples 5a and 5b, for instance. This means that ezafe information will help the model, and also humans, to better detect the phrase boundaries, which can be helpful in recognizing syntactic roles (Nourian et al, 2015).…”

Section: The Role Of Ezafementioning

confidence: 99%

Persian Ezafe Recognition Using Transformers and Its Role in Part-Of-Speech Tagging

Doostmohammadi

Nassajian

Rahimi³

2020

Findings of the Association for Computational Linguistics: EMNLP 2020

View full text Add to dashboard Cite

Ezafe is a grammatical particle in some Iranian languages that links two words together. Regardless of the important information it conveys, it is almost always not indicated in Persian script, resulting in mistakes in reading complex sentences and errors in natural language processing tasks. In this paper, we experiment with different machine learning methods to achieve state-of-the-art results in the task of ezafe recognition. Transformerbased methods, BERT and XLMRoBERTa, achieve the best results, the latter achieving 2.68% F 1 -score more than the previous stateof-the-art. We, moreover, use ezafe information to improve Persian part-of-speech tagging results and show that such information will not be useful to transformer-based methods and explain why that might be the case.

show abstract

Extending persian sentiment lexicon with idiomatic expressions for sentiment analysis

Dashtipour

Gogate

Gelbukh

et al. 2021

Soc. Netw. Anal. Min.

View full text Add to dashboard Cite

Nowadays, it is important for buyers to know other customer opinions to make informed decisions on buying a product or service. In addition, companies and organizations can exploit customer opinions to improve their products and services. However, the Quintilian bytes of the opinions generated every day cannot be manually read and summarized. Sentiment analysis and opinion mining techniques offer a solution to automatically classify and summarize user opinions. However, current sentiment analysis research is mostly focused on English, with much fewer resources available for other languages like Persian. In our previous work, we developed PerSent, a publicly available sentiment lexicon to facilitate lexicon-based sentiment analysis of texts in the Persian language. However, PerSent-based sentiment analysis approach fails to classify the real-world sentences consisting of idiomatic expressions. Therefore, in this paper, we describe an extension of the PerSent lexicon with more than 1000 idiomatic expressions, along with their polarity, and propose an algorithm to accurately classify Persian text. Comparative experimental results reveal the usefulness of the extended lexicon for sentiment analysis as compared to PerSent lexicon-based sentiment analysis as well as Persian-to-English translation-based approaches. The extended version of the lexicon will be made publicly available.

show abstract

On the Importance of Ezafe Construction in Persian Parsing

Cited by 10 publications

References 11 publications

FNLP‐ONT: A feasible ontology for improving NLP tasks in Persian

FNLP‐ONT: A feasible ontology for improving NLP tasks in Persian

Persian Ezafe Recognition Using Transformers and Its Role in Part-Of-Speech Tagging

Extending persian sentiment lexicon with idiomatic expressions for sentiment analysis

Contact Info

Product

Resources

About