A Text Reassembling Approach to Natural Language Generation
This addresses the problem of building NLG systems for domain experts lacking linguistics expertise, though it appears incremental as it builds on existing statistical techniques.
The paper tackles the drawbacks of existing statistical Natural Language Generation (NLG) methods, such as reliance on handcrafting and lack of transparency, by proposing a novel Text Reassembling approach (TRG) that is purely statistical and highly transparent, showing promise for domain experts with little NLG expertise.
Recent years have seen a number of proposals for performing Natural Language Generation (NLG) based in large part on statistical techniques. Despite having many attractive features, we argue that these existing approaches nonetheless have some important drawbacks, sometimes because the approach in question is not fully statistical (i.e., relies on a certain amount of handcrafting), sometimes because the approach in question lacks transparency. Focussing on some of the key NLG tasks (namely Content Selection, Lexical Choice, and Linguistic Realisation), we propose a novel approach, called the Text Reassembling approach to NLG (TRG), which approaches the ideal of a purely statistical approach very closely, and which is at the same time highly transparent. We evaluate the TRG approach and discuss how TRG may be extended to deal with other NLG tasks, such as Document Structuring, and Aggregation. We discuss the strengths and limitations of TRG, concluding that the method may hold particular promise for domain experts who want to build an NLG system despite having little expertise in linguistics and NLG.