Search or add a thesis

Advanced Search (Beta)
Home > Evaluation of Online Machine Translating Engines: A Comparative Study of English into Urdu Machine Translation

Evaluation of Online Machine Translating Engines: A Comparative Study of English into Urdu Machine Translation

Thesis Info

Author

Samina Tabassum

Department

Department of English

Program

Mphil

Institute

National University of Modern Languages

Institute Type

Public

City

Islamabad

Province

Islamabad

Country

Pakistan

Thesis Completing Year

2016

Subject

English Language

Language

English

Added

2021-02-17 19:49:13

Modified

2024-03-24 20:25:49

ARI ID

1676728735755

Asian Research Index Whatsapp Chanel
Asian Research Index Whatsapp Chanel

Join our Whatsapp Channel to get regular updates.

Similar


Evaluation of Online Machine Translating Engines: A Comparative Study of English into Urdu Machine Translation The study is about online machine translation systems and their comparative performance for English to Urdu language translation. Machine translation is an emerging field and is important to facilitate multilingual communication. In multilingual societies like Pakistan the demand for machine translation has increased sharply. Currently a number of online translators are providing automated translation services but do these tools provide accurate results and useful service is the subject that is under investigation. The current work is focused on evaluating the quality of four OMT systems, Google, Bing, Babylon and Worldlingo for English to Urdu translation through comparative analysis. Three types of source texts informative, expressive and operative written in English are selected by using Reiss’s (1997) text type classification (explained by Munday, 2001) for machine translation experimentation. The selected texts are experimented through each sample OMT engine and the obtained data is evaluated using the parameters and evaluation performa designed for the study. Help is taken from works of Hutchins and Somers (1992), Gaule and Josan (2012) and Halliday’s scale for accuracy and intelligibility used by Ellinder (2012) to design the theoretical framework and evaluation scale for the study. The study investigates the quality of sample machine translations through linguistic investigation, clarity evaluation, error analysis and comparative assessments. Evaluation process of the study is based on self perceived examination and 3rd person human evaluators’ observations. Using this collaborative evaluation technique quality of sample OMT engines is explored quantitatively through linguistic assessment at lexical and syntactic levels and through clarity assessment at semantic level. The strength and weaknesses of translation performance of sample OMT systems are also analyzed and discussed through error analysis of the experimented data. The obtained results from these three types of assessments are finally compared to develop arguments about performance of sample OMT engines. The results indicated marked differences in competency of sample OMT systems for English to Urdu translation.
Loading...
Loading...

Similar Books

Loading...

Similar Chapters

Loading...

Similar News

Loading...

Similar Articles

Loading...

Similar Article Headings

Loading...

لسانی تشکیلات اور افتخار جالب

موضوع11:لسانی تشکیلات اور افتخار جالب
لسانی تشکیلات کی تحریک یا روّیے کی مباحث افتخار جالب نے شروع کی۔ماضی کے لسانی ڈھانچے پر سوالات اٹھائے۔ اور لفظیات کے نئے آفاق کو ترتیب دیا۔ ان کا انسانی مسائل اور لسانی لہجے میں اظہاری انسلاک نظر نہیں آتا۔ جس میں شاعری کے ابہام ابھرتے ہیں۔ ان کے خیال میں ماضی اور اس سے متعلقہ تمام روایات اپنی موت آپ مرچکی ہے۔ ان کے خیال میں لغوی اور شعری معنئی کا فرق ایک تسلیم شدہ حقیقت ہے۔ لیکن افتخار جالب نے لغوی اور کشافی معنویت کے خلاف اپنا علم بغاوت اٹھا کر نعرہ قلندر بلند کیا کہ قواعد { گرائمر ، اجتماعیت اور ابلاغ تک توسیع دے دی جس سے چند مغاطے بھی پیدا ہوئے۔
افتخار جالب نے کلاسکیل اردو شاعری کے لسانی کے جمالیاتی اور اظہاری لیجے کے علاوہ ترقی پسند شعریاتی زبان پر سخت تنقید کی۔ حالانکہ افتخار جالب کا جھکاو بائیں بازو کی جانب تھا ، وہ ٹریڈ یونینسٹ تھے۔ اور ٹریڈ یونینسٹ سرگرمیوں میں فعال حصہ لینے کی پاداش میں ان کو الائید بنک کی نوکری سے ہاتھ دھونا پڑا۔ افتخار جالب دراصل نحوی ترکیب کا نیا شعری باطن تخلیق کرنا چاہتے تھے۔ انہوں نے معنی اور در معنی کے حوالے سے کئی سوالات اٹھائے۔ اور معنی میں معنی کو تلاش کرتے رہے۔ انہوں نے زنگ آلود افکار ، فرسودہ تراکیب، رموز، پامال کلمات، بدبو دار تشبہیات اور استعاروں کے خلاف آواز اٹھائی۔ اور انہوں نے اس جامد اسلوب شعر اور پھیکے جذبات کے خلاف آواز بلند کی۔ مگر ان کے قبیلے میں شامل شعرا ان کی بات کو نہیں سمجھ سکے اور ایک مضوعی دنیا کی شاعری کرتے رہے اور اردو میں ’’ نئی نئی شاعری’’ خلق نہیں کرسکے۔ اس سلسلے میں افتخار جالب نے اپنی کتاب ’’ لسانی تشکیلات اور قدیم بنجر پن میں...

Role of Religion in Mate Selection Among Educated Working Women in Khyber Pakhtunkhwa, Pakistan

The present study was conducted to probe the role of religion in mate selection among educated females of Khyber Pakhtunkhwa. Data for the study was collected through a pre-tested questionnaire from 278 working women from three public sector universities in Khyber Pakhtunkhwa i.e. The University of Agriculture, University of Peshawar and Shaheed Benazir Bhutto Women University Peshawar. Simple random method was applied to pick the samples from the study area whereas chi-square statistics was applied to get the major findings. Results reveal a contradictory type of situation as on the one hand, religion particularly the rituals were strictly followed during marriage but on the other hand, the mate selection rights of women were ignored mainly. This is practiced despite of the fact that Islam has given right/liberty in mate selection to both men and women.  Female respondents attributed it to lack of Islamic knowledge on part of family members or they do it to get more say and power through following common values and culture. Religious leaders do not lay much stress on the rights given by Islam to females in mate selection in their religious sermons. The issue according to respondents was not discussed either. Females on their part, though found interested in mates from their own sect or religious groups preferred liberal males over conservative ones because believing them to be stern and inflexible. The study recommends that an awareness campaign through mass media which may include religious leaders, civil society and political parties needs to be launched in order to raise awareness among both the educated and uneducated females along with their parents regarding mate selection, the nature and impact of religious misinterpretations, and current wave of western values, materialistic outlook and idealism that affects the selected females perception about mate selection.

A Disease Predictive Model for the Management of Chickpea Blight Ascochyta Rabiei Pass. Lab

Chickpea blight caused by Ascochyta rabiei (Pass.) Lab. (teleomorph: Didymella rabiei) (Kovachevski) v. Arx, is the most devastating disease of gram crop in the world. Disease can induce 100% yield losses under epidemic condition. In Pakistan, chickpea blight is causing heavy yield losses annually. Due to lack of resistance in indigenous chickpea germplasm, disease is controlled through fungicides by the growers of Pakistan. Excessive use of fungicides causes resistance in the pathogen and creates fatalistic effect on the environment. Chickpea blight disease predictive model under such situation may be effective tool to predict early onset of disease. In this way excessive use of fungicides may be avoided. Plant extracts and antagonists may also provide a replacement of chemical control of chickpea blight. The objective of this study was to develop a disease predictive model based upon environmental variables i.e. maximum and minimum temperatures, relative humidity, rainfall and wind speed to predict chickpea blight and devise an eco-friendly management strategy for this disease. Correlation and regression analysis was performed to determine the relationship of environmental variables with disease severity. Significant correlation was found between all environmental variables and disease severity. Maximum temperature showed negative correlation, while minimum temperature, relative humidity, rainfall and wind speed exhibited positive correlation with disease severity. Environmental factors and disease severity data of five years (2006-10) was used to develop a disease predictive model using stepwise regression analysis. Maximum and minimum temperatures, relative humidity, rainfall and wind speed significantly contributed in disease development and explained 72% variability in disease severity. This model, based on five years data, was then validated with two years (2011-12) environmental and disease severity data. In two years model all environmental parameters explained 82% variability in blight severity. Both models i.e. based upon five and two years data validated each other on the basis of homogeneity of regression lines. Blight severity was high at maximum (20-24°C) and minimum (12-14°C) temperatures, relative humidity (65-70%), rainfall (5-6 mm) and wind speed (5-6.5 km/h), respectively. Chickpea germplasm comprised of 48 genotypes was screened against chickpea blight during years 2011-12. Advanced lines exhibited resistant and moderately resistant response were viz; K-60013, K-98008, K-96001, K-96022, D-97092, D-91055, D-90272, D-96050, D-Pb2008 and D-Pu502-362, and K-96033, K-89169, K-90395, D-91017, D-89044, D-05006, D-96018, D-86030, D-96032, D-03009 and D-1CC-5127, respectively. For the management of disease, five fungicides, five plant extracts and two bio-control agents were evaluated under in vitro conditions. The concentrations of fungicides were 0.05%, 0.10% and 0.15%, respectively while concentrations of plant extracts were 3%, 5% and 8%, respectively. Spore suspension of bio-controls was kept as; 105, 106, and 107 conidia/mL, respectively. Means of treatments were compared using least significant difference (LSD) test. Two fungicides i.e. Alliete and ThiovetJet @ 0.15%, two plants extracts i.e. Melia azedarach and Azadirachta indica @ 8% and bio-control agent (Trichoderma harzianum at 107 conidia/mL) proved significantly effective under in vitro conditions. These most effective treatments were then applied under in vivo to check their efficacy for two years. Significant disease severity was reduced by fungicides i.e. Alliete (17%) and Thiovetjet (23%) followed by plant extracts, in which M. azedarach and A. indica reduced disease severity to 50% and 56%, respectively compared to control (75%). T. harzianum proved third good against chickpea blight disease in field after fungicides and plant extracts.