scholarly journals Corpus Quality Improvement to Improve the Quality of Statistical Translator Machines (Case Study of Indonesian Language to Java Krama)

2018 ◽  
Vol 1 (2) ◽  
pp. 65
Author(s):  
Muhammad Gerdy Asparilla ◽  
Herry Sujaini ◽  
Rudy Dwi Nyoto

Bahasa merupakan alat komunikasi yang dijadikan sarana untuk berinteraksi dengan masyarakat sekitar.Kemampuan akan penguasaan banyak bahasa tentunya akan mempermudah untuk berinteraksi dengan orang lain dari berbagai daerah yang berbeda. Oleh karena itu, diperlukan penerjemah untuk menambah pengetahuan akan berbagai bahasa yang ada. Mesin Penerjemah Statistik (Statistical Machine Translation) merupakan sebuah pendekatan mesin penerjemah dengan hasil terjemahan yang dihasilkan atas dasar model statistik yang parameter-parameternya diambil dari hasil analisis korpus paralel. Korpus paralel adalah pasangan korpus yang berisi kalimat-kalimat dalam suatu bahasa dan terjemahannya. Salah satu fitur yang digunakan untuk meningkatkan kualitas hasil terjemahan adalah dengan optimasi korpus. Tujuan yang ingin dicapai dalam penelitian ini adalah melakukan untuk melihat pengaruh kualitas korpus dengan memfilter pasangan kalimat-kalimat dengan terjemahan berkualitas. Filter yang digunakan adalah nilai minimal setiap kalimat yang di uji dengan metode Bilingual Evaluation Understudy (BLEU). Pengujian dilakukan dengan membandingkan nilai akurasi hasil terjemahan sebelum dan setelah optimasi korpus. Dari hasil penelitian, penggunaan optimasi korpus dapat meningkatkan kualitas terjemahan untuk mesin penerjemah bahasa Indonesia ke bahasa Jawa krama. Hal itu terlihat dari hasil pengujian dengan menambahkan optimasi korpus pada 15 kalimat uji diluar korpus terdapat peningkatan rata - rata nilai BLEU sebesar 10.53% dan dengan menggunakan 100 kalimat uji yang berasal dari korpus optimasi terdapat peningkatan rata-rata nilai BLEU sebesar 11.63%  pada pengujian otomatis serta 0.03% pada pengujian oleh ahli bahasa. Berdasarkan hal tersebut, mesin penerjemah statistik bahasa Indonesia ke bahasa Jawa krama dengan penggunaan fitur optimasi korpus dapat meningkatkan nilai akurasi hasil terjemahan.

2015 ◽  
Vol 8 (1) ◽  
pp. 1
Author(s):  
Afriantoni Afriantoni ◽  
Ibrahim Ibrahim

This study aimed to describe in depth between the link of school policy and the school quality improvement. The method in this study is a qualitative method using the case study presented descriptively. This research was conducted at SMA Negeri 2 Babat Tomat Kabupaten Musi Banyuasin. Based on this study it was found that the First, free school policy can help the economy / ease the burden of school costs to be incurred by the parents. Second, the policy constraints of the application for free school educa-tion at SMAN 2 Babat Toman is not very effective, so that the students' interest is not increasing, infrastructure is one of the obstacles in the implementation of free school education, how the quality of schools will be increased if it is not supported by facilities and complete infrastructure. Third, the quality of school education free SMAN 2 Babat Toman already realized well with regard to input, input turns unselected maximum, that is the students. Fourth, the implementation of free school education in Banyuasin, the quality of school SMAN 2 Babat Toman Muba Sumsel was not increased. This means that the implementation for free school education quality of school SMAN 2 Babat Toman was not increased.Keywords : free schools, school quality, case studies


Author(s):  
Hendri Wasito ◽  
Hening Pratiwi ◽  
Adi Wibowo ◽  
Nia Kurnia Solihat

Drugs are an important component of health services that are the needs of the community. There is still a lack ofcommunity knowledge of medicines and management especially for family members, hence an educational effort as well asimprovement of quality of drug management in family through training program and mentoring by pharmacist. Thiscommunity service activity aims to determine the knowledge and attitude of the community in managing drugs in the familyand improve the quality of drug management by the community in the family. The activity was conducted in SidasariWetanKubangkangkung Village Kawunganten Cilacap. The workshop on drug management in family was conducted by pharmaciststo the 33 participants. Data collection was done by using questionnaire and observation through home visit. The result of theactivity shows that the increase of knowledge and attitude of the society in managing drugs in the familywere 10% and 7%,respectively. Workshop activities and mentoring by pharmacists can provide benefits and behavioral changes in family drugsmanagement.


Author(s):  
Herry Sujaini

Extended Word Similarity Based (EWSB) Clustering is a word clustering algorithm based on the value of words similarity obtained from the computation of a corpus. One of the benefits of clustering with this algorithm is to improve the translation of a statistical machine translation. Previous research proved that EWSB algorithm could improve the Indonesian-English translator, where the algorithm was applied to Indonesian language as target language.This paper discusses the results of a research using EWSB algorithm on a Indonesian to Minang statistical machine translator, where the algorithm is applied to Minang language as the target language. The research obtained resulted that the EWSB algorithm is quite effective when used in Minang language as the target language. The results of this study indicate that EWSB algorithm can improve the translation accuracy by 6.36%.


2011 ◽  
Vol 95 (1) ◽  
pp. 87-106 ◽  
Author(s):  
Bushra Jawaid ◽  
Daniel Zeman

Word-Order Issues in English-to-Urdu Statistical Machine Translation We investigate phrase-based statistical machine translation between English and Urdu, two Indo-European languages that differ significantly in their word-order preferences. Reordering of words and phrases is thus a necessary part of the translation process. While local reordering is modeled nicely by phrase-based systems, long-distance reordering is known to be a hard problem. We perform experiments using the Moses SMT system and discuss reordering models available in Moses. We then present our novel, Urdu-aware, yet generalizable approach based on reordering phrases in syntactic parse tree of the source English sentence. Our technique significantly improves quality of English-Urdu translation with Moses, both in terms of BLEU score and of subjective human judgments.


2020 ◽  
Vol 8 (2) ◽  
pp. 137-145
Author(s):  
Hary Priatna Sanusi

This study was motivated by a phenomenon that appears in madrasa, namely: 1) the low quality of graduates, in MAN 1 with an average result of UAN 58.5 and MAN 2 57.8 from a maximum score of 100; 2) inconsistency in planning, implementing, evaluating programs to improve quality. The purpose of this study was to identify: work programs, program implementation, evaluation of program implementation, program impacts, and supporting factors inhibiting the improvement of the quality of madrasa. The approach of this study was qualitative-naturalistic, with a descriptive research type. This research method was a case study, with data collection techniques through interviews, observation, and documentation study. Meanwhile, the data analysis technique used the interactive Miles model. The results of this study showed: evaluation of quality improvement which is carried out through an evaluation of needs assessment, assessment of input, assessment of processes and assessment of results.


2018 ◽  
Vol 3 (2) ◽  
pp. 252
Author(s):  
Anang Taufik Karunia ◽  
Maswardi M. Amin ◽  
M. Chiar

<p>The purpose of this research is to understand and describe in detail and depth about 1) Planning of standard quality improvement process based on school-based management at State Junior High School 3 Singkawang (SMPN 3 Singkawang); 2) Organizing the improvement of process standard quality based on school-based management in SMP N 3 Singkawang; 3) Implementation of standard quality improvement process based on school-based management in SMP N 3 Singkawang; 4) Supervision of standard quality improvement process based on school-based management at SMP N 3 Singkawang; 5) obstacles and efforts in improving the quality of process standards based on School-based<br />management in SMPN 3 Singkawang. The approach used in this research is qualitative research using case study method. then the results obtained in this study as follows 1) Planning standard quality improvement process based on the<br />school-based management SMPN 3 Singkawang in good categorize; 2) Organizing the improvement of process standard quality based on school-based management of SMPN 3 Singkawang is categorized well; 3) Implementation of standard quality improvement process based on school-based management of SMPN 3 Singkawang is categorized quite well; 4) Supervision of standard quality improvement process based on school-based management of SMPN 3 Singkawang is categorized quite well; 5) Obstacles in improving the quality of process standards based on the school-based management of SMPN 3 Singkawang is the condition of facilities still not meet the standards, especially the land or building, the physical condition of teachers caused by the age factor but the quality improvement of educators developed through certification of educators and other activities that support the learning process as well as the establishment of harmonious relationships between schools, parents and alumni.</p>


2019 ◽  
Vol 28 (3) ◽  
pp. 447-453 ◽  
Author(s):  
Sainik Kumar Mahata ◽  
Dipankar Das ◽  
Sivaji Bandyopadhyay

Abstract Machine translation (MT) is the automatic translation of the source language to its target language by a computer system. In the current paper, we propose an approach of using recurrent neural networks (RNNs) over traditional statistical MT (SMT). We compare the performance of the phrase table of SMT to the performance of the proposed RNN and in turn improve the quality of the MT output. This work has been done as a part of the shared task problem provided by the MTIL2017. We have constructed the traditional MT model using Moses toolkit and have additionally enriched the language model using external data sets. Thereafter, we have ranked the phrase tables using an RNN encoder-decoder module created originally as a part of the GroundHog project of LISA lab.


2017 ◽  
Vol 25 (1) ◽  
pp. 5-25 ◽  
Author(s):  
Paul G. LeMahieu ◽  
Alicia Grunow ◽  
Laura Baker ◽  
Lee E. Nordstrum ◽  
Louis M. Gomez

Purpose The purpose of this paper is to delineate an approach to quality assurance in education called networked improvement communities (NICs) that focused on integrating the methodologies of improvement science with few of the networks. Quality improvement, the science and practice of continuously improving programs, practices, processes, products and services within organized social systems, is a still-evolving area in education. This paper is the first of seven elaborating upon different approaches to quality improvement in education[1]. It delineates a new methodology called the NICs model. Developed by the Carnegie Foundation for the Advancement of Teaching, the approach is aimed at continuously improving the quality of practices, processes and outcomes in targeted problem areas in education systems. Design/methodology/approach The paper presents the historical development, theoretical foundations, core principles and adaptation of key elements of the NICs model for quality improvement in education. A case study specifically examines the problem of fostering new teacher effectiveness and retention in large public school systems in the USA. Findings The six principles underlying the NICs model are as follows: make the work problem-specific and user-centered, focus on variation in performance, see the system that produces outcomes, improve at scale what you can measure, use disciplined inquiry to drive improvement and accelerate learning through networked communities. Originality/value Few theoretical treatments and demonstration cases are currently available that examine the application of common models of quality improvement in education. This paper elaborates on one promising approach. In addition to examining the NICs model, the paper derives added value by allowing comparisons with seven widely used quality improvement approaches treated in this volume.


Author(s):  
Haris Hidayat Ismail ◽  
M. Duskri ◽  
C. M. Zubainur ◽  
Said Munzir

The Programme International for Student Assessment (PISA) has become a tool in determining the quality of education in a certain country. Unfortunately, Indonesian students still have difficulties in solving PISA problems. In an effort to improve students' ability in Indonesia to solve PISA questions, we need to analyze their difficulties in solving contextual problems. In this study, we have developed PISA-like problems by taking into account the contextual environment of students and current Indonesian curriculum. Forty-eight PISA-like math problems in Bahasa Indonesia have been developed consisting of sixteen questions level one, sixteen questions level two, and sixteen questions level three. Then, those PISA-like math problems were used to assess the students’ ability (case study in SMPN 8 Banda Aceh, Indonesia). Our study revealed that the ability of students is very low in understanding the information given in the problems. The students’ ability in the mathematical process and decoding are also quite low. Students should practice the PISA-like problems in the teaching-learning process in order to improve their abilities in PISA.


Sign in / Sign up

Export Citation Format

Share Document