Bengt Sigurd

Deep comprehension, generation and translation of weather forecasts (weathra)

DEEP COMPREHENSION, GENERATION AND TRANSLATION OF WEATHER FORECASTS (WEATHRA) by BENGT SIGURD, CAROLINE WILLNERS, MATS EEG-OLOFSSON and CI-LRISTER JOHANSSON Dept of Linguistics, University of Lund, Sweden E-mail: linglund@gemini.ldc.lu.se FAX:46-(0)46 104210 Introduction and abstract Weather forecasts were early noted to be a domain where automatic translation was possible (Kittredge, 1973). Everybody in the field knows that there is a computer in Montreal translating forecasts routinely be- tween French and English (METEO). The weather domain has proven to be a fruitful domain for further research as witnessed e.g. by the system for generating marine forecasts presented by Kittredge et al (1986), by the work by Goldberg et al (1988), by the system gene- rating public weather reports in Bulgarian reported on by Mitkov (1991) and the system translating Finnish marine forecasts into Swedish by Bl~tberg (1988). The Swedish Weathra system to be presented in this paper explores the language and semantics of weather forecasts further and it aims at deep comprehension of forecasts. Beside grammatical re- presentations, Weathra uses repre- sentations of the meteorological raw facts and secondary facts, e.g. the fact that it will probably rain at a place where there is a low pressure area. It uses a representation of meteorological objects with their properties as frames in a data base and graphic representation with tile standard meteorological icons on a map, e.g. icons for sun, cloudy, rain, snow, thunderstorm, westerly winds, L(ow) and H(igh) pressure, temperatures, e.g. 10-15. Weathra also features a dynamic discourse representation including the discourse objects which may be referred to by the words and anaphora in the text (cf Karttunen, 1976, Johnson & Kay, 1990). The discourse objects are regarded as instances of the (proto)types or (concepts), which are also available as frames in a database. The formal grammar, morpho- logy and lexicon of Weathra are based on experience from the machine translation system Swetra (Sigurd & Gawronska, 1988), which is also written in Prolog (LPA MacProlog). The Weathra system can understand weather forecasts in a fairly deep sense, depict its comprehension in a map, answer questions about the main contents and consequences, translate English forecasts into Swedish ones and vice versa, and generate various forecast texts in English or Swedish.

A referent grammatical analysis of relative clauses

Abstract Relative clauses have traditionally been said to have an antecedent or a correlate corresponding to a coreferent, missing, and relativized constituent in the relative clause (some representative works on relative clauses are mentioned in the list of references). Referent grammar (Sigurd, 1987) assumes referent variables in the syntactic representations of noun phrases and it is natural to assume that it is a referent variable, which is the antecedent or the correlate, not an individual word or individual words. The advantages of this analysis will be shown in this paper. The paper includes a survey of the main types of relative clauses with some typological comments. Swedish and English will be used as the main languages of demonstration. As Referent Grammar (RG) is formalized in Definite Clause Grammar (DCG), a formalism supported by many Prolog programs, the analysis can be implemented directly and run on a computer. The analysis presented in this paper is part of a computer program used for au...

Modals as a problem for MT

S u m m a r y Tim paper demonstrates tim problem of translating modal verbs and phrases and shows how some of these problems can be overcome by choosing semantic representations which look like representations of passive verbs. These semantic representations suit alternative ways of expressing modality by e.g. passive constructions, adverbs and impersonal constructions in the target language. Various restructuring rules for English, Swe(lish and Russian am presented.

Non-finite auxiliaries as affixes

Abstract This paper presents an analysis of verb phrases, where non-finite auxiliaries are seen as affixes (in e.g. English and Swedish prefixes, in German and Japanese suffixes) to non-finite main verbs or (recursively) to verb complexes consisting of a main verb with affixed non-finite auxiliaries. The non-finite verb phrase in Bill may have wanted to begin to move is thus seen as a phrase where have has been prefixed to wanted to begin to move, wanted prefixed to to begin to move , and begin has been prefixed to to move. The infinitive marker to is considered as a kind of meaningless linking morpheme (glue) which some verbs require while others (the core modal auxiliaries) do not. The choice of non-finite form (infinitive, past participle or present participle) is also considered as a meaningless automatic consequence of the preceding verb. The auxiliary may requires an infinitive as its non-finite successor, want requires an infinitive, but with the infinitive marker to before it. The progressive be requires a present participle as its successor non-finite form as in Bill may be moving . The analysis has been implemented in the Prolog computer programs used in the machine translation system Swetra (Weathra) presented e.g. at Coling 92 (Sigurd et al. 1992) and the paper presents some demo printouts.

The passive as an auxiliary

Abstract The passive is one of the most intriguing grammatical phenomena and it has been discussed in innumerable articles and books. This paper will present the approach to the passive taken in the MT-projekt Swetra focusing the translation between Swedish and English (Sigurd et al. 1992). The approach to the passive may be said to be auxiliary-oriented (cf. Sigurd, 1992) and the suggested meaning representation of the passive looks the same as the representation of an auxiliary.