By Michael Carl, Srinivas Bangalore, Moritz Schaeffer
This quantity offers a entire advent to the interpretation method learn Database (TPR-DB), which was once compiled via the Centre for examine and Innovation in Translation and applied sciences (CRITT). The TPR-DB is a different source that includes greater than 500 hours of recorded translation approach facts, augmented with over 2 hundred diverse wealthy annotations. Twelve chapters describe the varied examine instructions this information can help, together with the computational, statistical and psycholinguistic modeling of human translation processes.
In the 1st chapters of this publication, the reader is brought to the CRITT TPR-DB. this is often through major components, the 1st of which makes a speciality of usability concerns and info of enforcing interactive desktop translation. It additionally discusses using exterior assets and translator-information interplay. the second one half addresses the cognitive and statistical modeling of human translation tactics, together with co-activation on the lexical, syntactic and discourse degrees, translation literality, and diverse annotation schemata for the data.
Read or Download New Directions in Empirical Translation Process Research: Exploring the CRITT TPR-DB PDF
Best data mining books
Data Mining in Agriculture represents a accomplished attempt to supply graduate scholars and researchers with an analytical textual content on info mining innovations utilized to agriculture and environmental similar fields. This booklet provides either theoretical and sensible insights with a spotlight on providing the context of every information mining process fairly intuitively with considerable concrete examples represented graphically and with algorithms written in MATLAB®.
This e-book comprises precious reports in information mining from either foundational and useful views. The foundational reports of information mining may also help to put an effective origin for facts mining as a systematic self-discipline, whereas the sensible stories of knowledge mining could lead to new information mining paradigms and algorithms.
This publication constitutes the refereed lawsuits of the seventeenth foreign convention on facts Warehousing and data Discovery, DaWaK 2015, held in Valencia, Spain, September 2015. The 31 revised complete papers offered have been conscientiously reviewed and chosen from ninety submissions. The papers are prepared in topical sections similarity degree and clustering; info mining; social computing; heterogeneos networks and knowledge; facts warehouses; circulate processing; purposes of huge info research; and massive facts.
This e-book is dedicated to the modeling and knowing of advanced city structures. This moment quantity of knowing complicated city structures specializes in the demanding situations of the modeling instruments, referring to, e. g. , the standard and volume of information and the choice of a suitable modeling method. it truly is intended to aid city decision-makers—including municipal politicians, spatial planners, and citizen groups—in making a choice on a suitable modeling strategy for his or her specific modeling requisites.
- Advances in Natural Language Processing: 9th International Conference on NLP, PolTAL 2014, Warsaw, Poland, September 17-19, 2014. Proceedings
- Advances in intelligent information and database systems
- Computer Vision - ECCV 2008: 10th European Conference on Computer Vision, Marseille, France, October 12-18, 2008, Proceedings, Part I
- Data Mining: Foundations and Intelligent Paradigms: Volume 3: Medical, Health, Social, Biological and other Applications
- Introduction to data mining and its applications
- Applied Data Mining for Business and Industry
Extra info for New Directions in Empirical Translation Process Research: Exploring the CRITT TPR-DB
For instance, an expression like “life sentences” could be aligned as a multi-word unit, or compositional as two different units. The number of source and target language words of the alignment unit (AU) of which “life” is part, is reflected in the SAUnbr and TAUnbr values respectively. The HSeg attribute takes into account this alignment segmentation context, and is calculated in a similar way as HTra with the difference that it relies on counting identical TAUnbr, instead of TToken. 4587 32 M.
MS13: This study is an investigation of translator’s behaviour when translating and post-editing Portuguese and Chinese in both language directions. 27. RH12: This is an authoring study for the production of news by two Spanish journalists. 28. ROBOT14: This study investigates usage of external resources during translation and post-editing. 2 The CRITT Translation Process Research Database 49 29. ZHPT12: This study investigates translator’s behaviour when translating journalistic texts. The specific aim is to explore translation process research while processing non-literal (metaphoric) expressions.
The more different equally probable translations a source word has, the higher is its word translation entropy H(s). Chapter 10, Sect. 2 in this volume gives a more in depth background on word translation entropy. Perplexity (PP) is related to entropy H, as an exponential function as shown in Eq. 4) The higher the perplexity, the more similarly likely choices exist and hence the more difficult is a decision to make. The ST tables provide some of this information: CountT represents the number of observed SToken !