loading subjects...
Ebook: Neural Generation of Textual Summaries from Knowledge Base Triples
loading...
Most people need textual or visual interfaces to help them make sense of Semantic Web data. In this book, the author investigates the problems associated with generating natural language summaries for structured data encoded as triples using deep neural networks.
An end-to-end trainable architecture is proposed, which encodes the information from a set of knowledge graph triples into a vector of fixed dimensionality, and generates a textual summary by conditioning the output on this encoded vector. Different methodologies for building the required data-to-text corpora are explored to train and evaluate the performance of the approach. Attention is first focused on generating biographies, and the author demonstrates that the technique is capable of scaling to domains with larger and more challenging vocabularies.
The applicability of the technique for the generation of open-domain Wikipedia summaries in Arabic and Esperanto – two under-resourced languages – is then discussed, and a set of community studies, devised to measure the usability of the automatically generated content by Wikipedia readers and editors, is described.
Finally, the book explains an extension of the original model with a pointer mechanism that enables it to learn to verbalise in a different number of ways the content from the triples while retaining the capacity to generate words from a fixed target vocabulary. The evaluation of performance using a dataset encompassing all of English Wikipedia is described, with results from both automatic and human evaluation both of which highlight the superiority of the latter approach as compared to the original architecture.
Most people need textual or visual interfaces in order to make sense of Semantic Web data. In this thesis, we investigate the problem of generating natural language summaries for structured data encoded as triples using deep neural networks.
We propose an end-to-end trainable architecture that encodes the information from a set of triples into a vector of fixed dimensionality and generates a textual summary by conditioning the output on this encoded vector. In order to both train and evaluate the performance of our approach, we explore different methodologies for building the required data-to-text corpora. We initially focus our attention on the generation of biographies. Using methods for both automatic and human evaluation, we demonstrated that our technique is capable of scaling to domains with challenging vocabulary sizes of over 400k words.
Given the promising results of our approach in biographies, we explore its applicability in the generation of open-domain Wikipedia summaries in two under-resourced languages, Arabic and Esperanto. We propose an adaptation of our original encoder-decoder architecture that outperforms a set of strong baselines of different nature. Furthermore, we conducted a set of community studies in order to measure the usability of the generated content by Wikipedia readers and editors. The targeted communities ranked our generated text close to the expected standards of Wikipedia. In addition, we found that the editors are likely to reuse a large portion of the generated summaries, thus, emphasizing the usefulness of our approach to the involved communities.
Finally, we extend the original model with a pointer mechanism that enables it to jointly learn to verbalise in a different number of ways the content from the triples while retaining the ability to generate regular words from a fixed target vocabulary. We evaluate performance with a dataset encompassing the entirety of English Wikipedia. Results from both automatic and human evaluation highlight the superiority of the latter approach compared to our original encoder-decoder architecture and a set of competitive baselines.