Table-of-Contents Generation on Contemporary Documents

Bentabet, Najah-Imane; Juge, Rémi; Ferradans, Sira

doi:10.1109/icdar.2019.00025

Cited by 5 publications

(2 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As we will see in section "'Experimental datasets', there are multiple ways in which section nesting can be signaled in documents using font style or numbering and a depth level of 5 generally applies to a large set of document types. This paper extends [28]. The main novelty is the introduction of a new dataset in the financial domain consisting of TOC annotations of English-written investment documents, allowing us to run an extensive analysis of our models results.…”

Section: Paper Contributionsmentioning

confidence: 96%

“…The baselines, as well as the proposed method, are quantitatively assessed against this new dataset. Another difference with [28] lies in the implementation of Rahman's and Finin model [29] which stands as our strong baseline. We use the code made available by the authors instead of rewriting their model ourselves (as we did originally) for fairer comparison.…”

Section: Paper Contributionsmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Table-of-Contents Generation for Efficient Information Access

et al. 2020

View full text Add to dashboard Cite

Purpose This paper presents a novel neural-based approach, applicable to any searchable PDF document that first detects the titles and then hierarchically orders them using a sequence labelling approach to generate automatically the Table of Contents (TOC). A TOC signals the main divisions and subdivisions of a document to assist with navigation and information localisation. Methods Unlike previous methods, we do not assume the presence of parsable TOC pages in the document but infer the TOC from a data-driven analysis of sections titles, their order and their depth. Results We offer an exhaustive analysis of the proposed model and evaluate it on French and English using documents from the financial domain, which we release to increase community's interest. We compare this model to state-of-the-art approaches and show its superiority in multiple experiments. Conclusions The approach described in this paper can easily be adapted to other domains and documents and its application to the analysis of financial prospectuses will be strengthened by the release of datasets. The TOC generation algorithms used in this paper obtain state-of-the-art results and provide strong baselines for future work.

show abstract

Section: Paper Contributionsmentioning

confidence: 96%

Section: Paper Contributionsmentioning

confidence: 99%