MFMohammad Reza Falahati Qadimi Fumani

Mohammad Reza Falahati Qadimi Fumani

*This is an algorithmically generated author record based on publicly available information.

Mohammad Reza Falahati Qadimi Fumani has a total of 7 co-authors. Here are the top co-authors by number of shared publications:

  • Hediye Mansouri

    M.A. in English Language Translation, Freelance Translator

    2 shared publications
  • Bahareh Pahlevanzadeh

    Faculty Member of the Department of System Operations, Islamic World Science and Technology Monitoring and Citation Institute, Shiraz, Iran

    1 shared publication
  • C. S. Ramachandra

    Reader in Linguistics, Dept. of Studies in Linguistics, Univ. of Mysore, India

    1 shared publication
  • C.S. Ramachandra

    Department of Studies in Linguistics, Kuvempu Institute of Kannada Studies, University of Mysore, Mysore, India; Tel: 0091 821 2545294

    1 shared publication
  • Ehsan Amouee

    MSc from the Department of Computational Linguistics, Islamic World Science and Technology Monitoring and Citation Institute, Shiraz, Iran

    1 shared publication
  • Marzieh Goltaji

    Islamic World Science Citation Center

    1 shared publication

Latest Publications

  • Open Access

    Dependency Parsing in Persian: A Rule-based Hybrid Model

    by Ehsan Amouee, Mohammad Reza Falahati Qadimi Fumani, Bahareh Pahlevanzadeh

    Dependency Parsing in Persian is mostly data-driven. In this paper, we have presented a hybrid model of grammar-based parsing method consisting of Hyperedge Replacement and Constraint Dependency Grammars. The model parses Persian POS tagged sentences based on their main verb valency structure in a rule-based manner. The sample data of the study, selected through purposive sampling, included 81 Persian sentences covering all types of Persian basic sentence structures. The results of the comparison of each sentence in Gold Analysis and the corresponding output of the model with both UAS and LAS have been presented in Table 10. These scores pertain to the entire research corpus and have been calculated based on the average score of each sentence result. HRG is fully capable of representing all base sentence structures as hypergraphs and in a sentence, could generate all possibilities for any desired sequence of POS tags. In our hybrid model, HRG collaborated with CDG on dependency parsing, and with the assistance of the heuristic function, the model could achieve an accuracy of 100% among the research corpus. Further, the represented model could undertake dependency parsing with no train dataset. Also, we presented a machine-readable verb valency dictionary, based on our proposed HRG model and corpus data. Although the model’s time complexity is quasi-polynomial time, the accuracy score in our sample data might be helpful.

  • Translation Students’ Editing Ability: The Case of IAPLL Guidelines

    by Hediye Mansouri, Mohammad Reza Falahati Qadimi Fumani

    This study intended to investigate the ability of B.A. and M.A. students of English language translation in identification and correction of errors in a sample Persian text based on Iran’s Academy of Persian Language and Literature’s editing guidelines. An error-embedded researcher-made text was administered to the students to assess their error identification and correction ability. Using whole group sampling, two groups of 35 B.A. level and 35 M.A. students of English language translation were used as participants. The results indicated that the mean of the total score in B.A. group was significantly lower than the medium level; the mean of the total score in M.A. group was at the medium level; recognition score of the M.A. group was significantly higher than the B.A. group; correction score of the M.A. group was significantly larger than the B.A. group, and the total score of the M.A. group was significantly larger than the B.A. group. In general, it was found that the participants’ editing ability was very low. This could have many reasons including lack of qualitative teaching or courses on IAPLL rules and regulations in English language translation syllabus. The findings of this study are in line with findings reported by Zare and Falahati Qadimi Fumani (2020) who reported that students of English language translation in Iran were weak at editing skills and specifically know very little about the terms approved by IAPLL.

  • Translation Students’ Editing Ability: The Case of IAPLL Guidelines

    by Hediye Mansouri, Mohammad Reza Falahati Qadimi Fumani

    This study intended to investigate the ability of B.A. and M.A. students of English language translation in identification and correction of errors in a sample Persian text based on Iran’s Academy of Persian Language and Literature’s editing guidelines. An error-embedded researcher-made text was administered to the students to assess their error identification and correction ability. Using whole group sampling, two groups of 35 B.A. level and 35 M.A. students of English language translation were used as participants. The results indicated that the mean of the total score in B.A. group was significantly lower than the medium level; the mean of the total score in M.A. group was at the medium level; recognition score of the M.A. group was significantly higher than the B.A. group; correction score of the M.A. group was significantly larger than the B.A. group, and the total score of the M.A. group was significantly larger than the B.A. group. In general, it was found that the participants’ editing ability was very low. This could have many reasons including lack of qualitative teaching or courses on IAPLL rules and regulations in English language translation syllabus. The findings of this study are in line with findings reported by Zare and Falahati Qadimi Fumani (2020) who reported that students of English language translation in Iran were weak at editing skills and specifically know very little about the terms approved by IAPLL.

  • The Persian Agrovoc in an indexing context 1

    by Mohammad Reza Falahati Qadimi Fumani

    The author explores the Persian Agrovoc thesaurus from the point of view of the indexer, both human and automatic. It is found wanting, with problems both in the English version and in the Persian translation.


  • Inter-indexer consistency (IIC) in a Persian context

    by Mohammad Reza Falahati Qadimi Fumani

    The author describes research he has carried out to assess the consistency (IIC) between pairs of indexers working on a collection of 111 Persian articles, using the Hooper, Rolling and cosine evaluation measures. By and large the IIC scores have been very low. He asks why this should be so, and to remedy the problem recommends better selection and training of indexers, better work conditions, and a collaborative approach to the task. He also points to the possibility that automatic indexing may offer a way forward.


  • Can an English Word Frequency Counter Serve the Requirements of Persian Automatic Indexing?

    by Mohammad Reza Falahati-Fumani, C.S. Ramachandra

    This article assessed the possible suitability of an English word frequency counter for Persian. Ten Persian articles related to agriculture were input to the software. The error analysis revealed that it had the potential to be used for Persian but modifications were necessary to reduce the errors. The main source of the errors lied in Persian morphology and alphabet and the availability of Persian and Arabic fonts, which permitted writers to introduce unsystematic variations in their writings. It was suggested that the Persian text be handled with all its inconsistencies through algorithms that would merge variations before the frequency count in run, or through deleting illegal spaces and converting them into full spaces and then using a closed list of bound morphemes to join the elements of multi unit words together.