Align then Summarize: Automatic Alignment Methods for Summarization Corpus Creation

Abstract

Summarizing texts is not a straightforward task. Before even considering text summarization, one should determine what kind of summary is expected. How much should the information be compressed? Is it relevant to reformulate or should the summary stick to the original phrasing? State-of-the-art on automatic text summarization mostly revolves around news articles. We suggest that considering a wider variety of tasks would lead to an improvement in the field, in terms of generalization and robustness. We explore meeting summarization: generating reports from automatic transcriptions. Our work consists in segmenting and aligning transcriptions with respect to reports, to get a suitable dataset for neural summarization. Using a bootstrapping approach, we provide pre-alignments that are corrected by human annotators, making a validation set against which we evaluate automatic models. This consistently reduces annotators' efforts by providing iteratively better pre-alignment and maximizes the corpus size by using annotations from our automatic alignment models. Evaluation is conducted on public_meetings, a novel corpus of aligned public meetings. We report automatic alignment and summarization performances on this corpus and show that automatic alignment is relevant for data annotation since it leads to large improvement of almost +4 on all ROUGE scores on the summarization task.

Councils in Action: Automating the Curation of Municipal Governance Data for Research

“…A more general form of such work might look to measure the topical and legislative diffusion across the country, for example answering the question: "how long does it take for similar legislative actions regarding a topic to occur in multiple different municipalities?" There are additional computational research questions available for investigation with the Councils in Actions dataset such as the research and development of methods for minutes items and transcript alignment or even more generally, models for "outline generation" to automatically generate the minutes items of a meeting from a transcript (Tardy et al, 2020, Zhang et al, 2019.…”