Word Completion- A First Step Toward Target-Text Mediated IMT

Word Completion: A First Step Toward Target-Text Mediated IMT George Foster, Pierre Isabelle, and Pierre Plamondon Centre for hfformation Technology Innovation (CITI) 1575 Chomedey Blvd. Laval, Quebec, Canada, H7V 2X2 {foster, isabelle, plamondon}@citi, doc. ca Abstract We argue that the conventional approach to Interactive Machine ~Ih-anslation is not the best way to provide assistance to skilled translators, and propose an alter- native whose central feature is the use of the target text as a medium of inter- action. We describe an automatic word- completion system intended to serve as a vehicle for exploring the feasibility of this new approach, and give results in terms of keystrokes saved in a test corpus. 1 Introduction Machine translation is usually significantly infe- rior to human translation, and for most appli- cations where high-quality results are needed it must be used in conjunction with a human trans- lator. There are essentially three ways of organiz- ing the process by which a person and a machine cooperate to produce a translation: prccdition, in which the person's contribution takes the form of a source-text analysis and occurs before the MT system is brought to bear; postedition, in which the translator simply edits the system's output; and interactive MT (IMT), which involves a di- alog between person and machine. Of the three, IMT is the most ambitious and theoretically the most powerflfl. It has a potential advantage over postedition in that information imparted to the system may help it to avoid cascading errors that would later require much greater effort to correct; and it has a potential advantage over preedition in that knowledge of the machine's current state may be useful in reducing the number of analyses the human is required to provide. Existing approaches to IMT (Blanchon, 1994; Boitet, 1990; Brown and Nirenburg, 1990; Kay, 1973; Maruyama and Watanabe, 1990; Whitelock et al., 1986; Zajac, 1988) place the MT system in control of the translation process and for the most part limit the human's role to performing various source language disambiguations on de- man& Although this arrangement is appropriate for some applications, notably those in which the user's knowledge of tile target language may be limited, or where there are multiple target, lan- guages, it is not well suited to tile needs of pro- fessional oi' other highly skilled translators. The lack of direct human control over the tinal target text (modulo postedition) is a serious drawback in this case, and it is not clear that, for a com- petent translator, disambiguat, ing a source text, is much easier than translating it. This conclusion is supported by the fact that true IMT is not, to our knowledge, used in most modern transla- tor's support environments, eg (Eurolang, 1995; I,'rederking et al., 1993; IBM, 1995; Kugler et al., 1991; Nirenburg, 1992; ~li'ados, 1995). Such envi- ronments, when they incorporate MT at all, tend to do so wholesale, giving the user control over whether and when an MT component is invoked, as well as extensive postediting facilities for …

Content truncated to 500 words