ABSTRACT. The main stream in machine translation is to build systems that are able to translate everything, but without any guarantees of quality. An alternative to this is systems that aim at precision but have limited coverage. Combining wide coverage with high precision is considered unrealistic. Most wide-coverage systems are based on statistics, whereas precision-oriented domain-specific systems are typically based on grammars, which guarantee translation equality by some kind of formal semantics.

This talk introduces a technique that combines wide coverage with high precision, by embedding a high-precision semantic grammar inside a wide-coverage syntactic grammar, which in turn is backed up by a chunking grammar. The system can thus reach good quality whenever the input matches the semantics; but if it doesn't, the user will still get a rough translation. The levels of confidence can be indicated by using colours, whence the title of the talk.

The talk will explain the main ideas in this technique, based on GF (Grammatical Framework) and also inspired by statistical methods (probabilistic grammars) and the Apertium system (chunk-based translation), boosted by freely available dictionaries (WordNet, Wiktionary), and built by a community of over 50 active developers. The current system covers 11 languages and is available both as a web service and as an Android application.

Demo system: http://www.grammaticalframework.org/demos/translation.html

ABSTRACT. In industry and academics alike, there is a renewed interest in data processing in general, and in deriving value from widely available data more specifically. The applications developed to this end often have non-experts as audience, which has led to a lot of research into verbalizing data into natural language. However, the proposed solutions often include predictive technologies that, along with predicted future events or behaviours, also generate additional information such as probabilities and confidence scores. This gives rise to data that goes beyond future facts alone and that has to be communicated to its users in order to enable a correct interpretation of the predictions. In this paper we explore the additional natural language needed when applications offer predictive analytics technology to their users. We demonstrate how to modularly implement the grammars needed to verbalize predictive aspects, independent of (existing) conceptualizations and lexical resources available for the domain of the data.

ABSTRACT. Knowledge and reasoning in a specific subject domain can greatly assist in natural language processing. This is demonstrated in the context of a question answering system for accessing data for business enterprise software.