Education, Science, Technology, Innovation and Life
Open Access
Sign In

Machine Translation Quality Estimation Algorithm Based on Intelligent Fuzzy Decision Tree Algorithm

Download as PDF

DOI: 10.23977/acss.2022.060709 | Downloads: 14 | Views: 408

Author(s)

Ruichao Li 1

Affiliation(s)

1 School of Translation Studies, Xi'an Fanyi University, Xi'an, Shaanxi, China

Corresponding Author

Ruichao Li

ABSTRACT

With the development and application of machine translation(MT) technology, MT results appear in more scenarios, but the translation quality cannot be guaranteed, and users need to know the quality of MT results to decide whether to adopt them or not. MT quality estimation (QE) is a key task in the field of MT, which can score the quality of a translation based only on the source language sentences and the MT. Unlike the methods of automatic translation evaluation, translation QE does not require the use of reference translation, which can save a lot of manpower and resources and is suitable for large-scale MT quality assessment scenarios without reference translation. In this paper, a phrase-structured syntactic tree is constructed based on an intelligent fuzzy decision tree(DT) algorithm, and the sentence-level and word-level QE results of MT translation are analyzed based on this tree structure. By comparing the training prediction results of translation QE models, it is obtained that the method of fusing dependent words with source language(SL) sequences is more helpful to the translation QE process than the method of fusing phrase structure features with source language sequences.

KEYWORDS

Machine Translation, Intelligent Fuzzy Decision Tree Algorithm, Translation Quality Estimation, Phrase Structure Syntax Tree

CITE THIS PAPER

Ruichao Li, Machine Translation Quality Estimation Algorithm Based on Intelligent Fuzzy Decision Tree Algorithm. Advances in Computer, Signals and Systems (2022) Vol. 6: 60-66. DOI: http://dx.doi.org/10.23977/acss.2022.060709.

REFERENCES

[1] Binh Nguyen, Binh Le, Long H. B. Nguyen, Dien Dinh: PhraseAttn. (2022) Dynamic Slot Capsule Networks for Phrase Representation in Neural MT. J. Intell. Fuzzy Syst, 4, 3871-3878.
[2] Irene Rivera-Trigueros. (2022) MT Systems and Quality Assessment: A Systematic Review. Lang. Resour. Evaluation, 2, 593-619.
[3] Sainik Kumar Mahata, Avishek Garain, Dipankar Das, Sivaji Bandyopadhyay. (2022) Simplification of English and Bengali Sentences for Improving Quality of MT. Neural Process. Lett. 4, 3115-3139.
[4] Arda Tezcan, Veronique Hoste, Lieve Macken. (2020) Estimating Word-level Quality of Statistical MT Output Using Monolingual Information Alone. Nat. Lang. Eng, 1, 73-94.
[5] Jani Dugonik, Borko Boskovic, Janez Brest, Mirjam Sepesy Maucec. (2019) Improving Statistical MT Quality Using Differential Evolution. Informatica, 4, 629-645.
[6] Sahinur Rahman Laskar, Abdullah Faiz Ur Rahman Khilji, Partha Pakray, Sivaji Bandyopadhyay.  (2022) Improved Neural MT for Low-resource English-Assamese Pair. J. Intell. Fuzzy Syst, 5,4727-4738.
[7] Ahmad Alos, Zouhair Dahrouj. (2020) Decision Tree Matrix Algorithm for Detecting Contextual Faults in Unmanned aerial Vehicles. J. Intell. Fuzzy Syst, 4,4929-4939.
[8] Subhashini Narayan, Jagadeesh Gobal. (2019) Optimal Decision Tree Fuzzy Rule-based Classifier for Heart Disease Prediction Using Improved Cuckoo Search Algorithm. Int. J. Bus. Intell. Data Min,4,408-429.
[9] K. Ramya, Yuvaraja Teekaraman, K. A. Ramesh Kumar. (2019) Fuzzy-Based Energy Management System With Decision Tree Algorithm for Power Security System. Int. J. Comput. Intell. Syst,2,1173-1178.
[10] Barry Haddow, Rachel Bawden, Antonio Valerio Miceli Barone, Jindrich Helcl, Alexandra Birch.(2022) Survey of Low-Resource MT. Comput. Linguistics, 3, 673-732.
[11] Christian J. Park, Paul H. Yi, Hussain Al Yousif, Kenneth C. Wang. (2022) Machine vs. Radiologist-Based Translations of RadLex: Implications for Multi-language Report Interoperability. J. Digit. Imaging, 3, 660-665.
[12] Rajesh Kumar Chakrawarti, Jayshri Bansal, Pratosh Bansal.(2022) MT Model for Effective Translation of Hindi Poetries into English. J. Exp. Theor. Artif. Intell, 3,95-109.

Downloads: 11339
Visits: 240416

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.