Cross-Domain Cue Switching Tiansi Dong tdong@uni-bonn.de AI Foundations Group Bonn-Aachen International Center for Information Technology (B-IT) University of Bonn
Translation Shared Knowledge in Spatial Description Approach to Shared Knowledge between Languages
Translation
Language Translation Example 1 How shall we translate white as snow into a native language in Benin (Natemba)?
white as snow very white white as pelican
Language Translation Example 2 How shall we translate you are my heart into Indonesian?
the heart is the most you are my heart important organ you are the most important person to me the liver is you are my liver the most important organ
In different languages, the same meaning may be carried by words w i t h t o t a l ly d i f fe r e n t l i t e r a l meanings. Translation needs speakers to select the right word in the target language which may have a distorted meaning structure from the source language
Language Translation Example 3 The Guugu Yimidhirr people only use absolute orientations: „on the southern edge of the western table“. How shall we translate it into the normal English expression?
N I left it on the western table It is on the table on my left side
North I left it on the western table West East ß South
It is on the table on my left Left side ß
West West left ß
� � Shared Knowledge between Orientation Descriptions the western Select: West Pole table Shared Knowledge: Newtonian Distance Comparison perspective the table Select: Heart on my left side Lebnizian Distortion perspective
Shared Knowledge among all Orientation Descriptions The distance comparison P: Ae i θ W The orientation of P is the point Q on the unite circle which is nearest to P.
Shared Knowledge among all Orientation Descriptions T.Dong and H.Guesgen „A Uniform Framework for Orientation Relations based on Distance Comparison“ ICCI’08. Distort the unite circle into a polygon front Define a distance function left right between point and a line segment behind Generate all Qualitative orientation framework in the literature
Distance Relations In UK, Object A is one foot away from Object B. B A In Germany, people use double-feet, or Elle In ancient Egypt, China, people use Cubit, Chi( 尺 ), Cun( ⼨对 ) Physicists use light: light-year is the distance travelled by light in one year; one meter is is the distance of light in vacuum in 0.00000000333564 second
Shared Knowledge among Distance Relations The connection relation Select an object category (shape) as unit The minimal number of elements in the object category, which can connect with the two objects, with the condition that they are connected with each other A B
J. Piaget Connection Relation S. Carey T.Dong „A Comment on RCC—from RCC to RCC ++ “, Journal of Philosophical Logic ∀ A, B [ C ( A, B ) → ∀ Z ∃ Z ∈ Z [ C ( A, Z ) ∧ C ( B, Z )]] B A W V
Shared Knowledge for Spatial Relations (Summary) absolute qualitative orientation orientation Select Distort Distance Distance comparison Connection
Shared Knowledge in Spatial Descriptions A table is under the projector A projector is on the table Commonsense knowledge of stability A tree leans against the bike A bike leans against the tree
Shared Knowledge in Spatial Domain T.Dong „Recognizing Variable Environment-The Theory of Cognitive Prism“, Springer Suppose two persons in the same environment, what is their shared spatial knowledge? Object classification at basic-level category Commonsense knowledge of stability Object locations based on the connection relation Supported by the case of L.E.
Why Are W e Interested in Spatial Domain? The first Domain that human encounters and understands It is the reference domain for the cognition of other domains A common research field for several disciplines
Translation without understanding can be incorrect, sometimes impossible Descriptions about spatial domain in different languages or cultures have shared knowledge What is the knowledge shared between languages? Approach to Shared Knowledge between Languages
The Shared Knowledge between Languages of Bilingual Speakers MacWhinney, B., Bates, E., & Kliegl, R. (1984). „Cue validity and sentence interpretation in English, German, and Italian“. Journal of Verbal Learning and Verbal Behavior, 23:127-150. The Aggregated Language Model Cue Cue Separation between Meanings and Forms.
Cue-Switching for Translation Produce form Retrieve meaning by Cues of L2 by Cues of L1 cue switching 26
Five Basic Cues in Language Cue: the relation between form and meaning Five cues found in 165 Languages 1) Agreement German: 1) + 3) e.g. you are, I am 2) Animacy English: 2) + 5) e.g. a dog bites bones 3) Case Chinese: 4) + 5) e.g. ich, mich, mir; I, me 4) Marker Turkey: 3) e.g. 把，被，着，了 Italian: 1) 5) Word order 27 e.g. a dog bites a cat
W ord Salat Problem 165 Natural Languages with only 5 cues Meaning construction with understandable, but ungrammatical sentences There should be some algorithms, software framework to construct possible meanings only based on single word meanings.
����� ������ ������� German-Chinese Translation based on Cue Switching Frame: gefallen Das Auto gefällt mir gut cues Subject: das Auto Frame: gefallen Receptor: mir used in Subject Receptor Descriptor: gut German I:mir: 我 the car: das Auto: 这⻋轧 like: gefällt: 喜欢 well:gut: 很 Frame: 喜欢 Actor Object cues Actor: 我 我很喜欢这⻋轧 Object: 这⻋轧 Frame: 喜欢 used in Descriptor: 很 Chinese
Translation based on Cue Switching supported by Shared Knowledge Frame: On X on the NRST(table,W-Pole) Object: X western Location: table cues table[West(table)] ι used in NRST(I.Front,N-Pole) English C(I.Left, W-Pole) visual, On X 在 .. 上 GPS, table 桌上 dialog West ⻄覀⾯靣 NRST(table,I.Left) / 桌上 ⻄覀⾯靣 ∈ dom West I.Left I.Left 我左边 X 在我左 Frame: 在 .. 上 边的桌上 cues Object: X used in Location: Chinese 桌上 [ 我左边 ( 桌上 )] ι
verbal gesture cue cue facial Each cue provides auditory cue knowledge from one cue perspective about the physical world and/or the psychological world. General principles of the two worlds haptic cue cue voice make the transition possible cue cue olfactory visual
Schematization Similarity Conjecture To the extent that space is schematized similarly in language and cognition, language will be successful in conveying space. — B. Tversky & P. Lee
On-going Tasks Acquire Shared Meaning Representation of German and Chinese sentences in textbooks Developing Software Systems for Machine Translation based on Cue-Switching Developing algorithms to solve for Word Salad Problem supported by DBpedia
Thanks
Recommend
More recommend
