{"id":{"repo_id":"lethbridge","oai_identifier":"oai:opus.uleth.ca:10133/5827"},"canonical_url":"https://search.dev.ndltd.org/etd/lethbridge/oai:opus.uleth.ca:10133/5827","repository":{"repo_id":"lethbridge","name":"University of Lethbridge","base_url":"https://opus.uleth.ca/server/oai/request"},"display":{"title":"Abstractive multi-document summarization - paraphrasing and compressing with neural networks","abstract":"This thesis presents studies in neural text summarization for single and multiple documents.The focus is on using sentence paraphrasing and compression for generating fluent summaries, especially in multi-document summarization where there is data paucity. A novel solution is to use transfer-learning from downstream tasks with an abundance of data. For this purpose, we pre-train three models for each of extractive summarization, paraphrase generation and sentence compression. We find that summarization datasets – CNN/DM and NEWSROOM – contain a number of noisy samples. Hence, we present a method for automatically filtering out this noise. We combine the representational power of the GRU-RNN and TRANSFORMER encoders in our paraphrase generation model. In training our sentence compression model, we investigate the impact of using different early-stopping criteria, such as embedding-based cosine similarity and F1. We utilize the pre-trained models (ours, GPT2 and T5) in different settings for single and multi-document summarization.","abstract_html":"This thesis presents studies in neural text summarization for single and multiple documents.The focus is on using sentence paraphrasing and compression for generating fluent summaries, especially in multi-document summarization where there is data paucity. A novel solution is to use transfer-learning from downstream tasks with an abundance of data. For this purpose, we pre-train three models for each of extractive summarization, paraphrase generation and sentence compression. We find that summarization datasets – CNN/DM and NEWSROOM – contain a number of noisy samples. Hence, we present a method for automatically filtering out this noise. We combine the representational power of the GRU-RNN and TRANSFORMER encoders in our paraphrase generation model. In training our sentence compression model, we investigate the impact of using different early-stopping criteria, such as embedding-based cosine similarity and F1. We utilize the pre-trained models (ours, GPT2 and T5) in different settings for single and multi-document summarization.","abstract_has_math":false,"creators":["Egonmwan, Elozino Ofualagba","University of Lethbridge. Faculty of Arts and Science"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2020,"date_issued":"2020","date_published":"2020","updated_at":"2026-07-27T20:02:41Z","subjects":["Artificial intelligence","Automatic programming (Computer science)","Computer programming","Dissertations, Academic","Machine learning","Natural language processing (Computer science)","Neural networks (Computer science)"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["hdl:10133/5827"],"render_values":[{"text":"hdl:10133/5827","href":null,"code":true}]}]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null},"metadata_groups":[{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.issued","label":"Date","values":["2020"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Artificial intelligence","Automatic programming (Computer science)","Computer programming","Dissertations, Academic","Machine learning","Natural language processing (Computer science)","Neural networks (Computer science)"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["hdl:10133/5827"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.other","label":"Dc Description Other","values":["This thesis presents studies in neural text summarization for single and multiple documents.The focus is on using sentence paraphrasing and compression for generating fluent summaries, especially in multi-document summarization where there is data paucity. A novel solution is to use transfer-learning from downstream tasks with an abundance of data. For this purpose, we pre-train three models for each of extractive summarization, paraphrase generation and sentence compression. We find that summarization datasets – CNN/DM and NEWSROOM – contain a number of noisy samples. Hence, we present a method for automatically filtering out this noise. We combine the representational power of the GRU-RNN and TRANSFORMER encoders in our paraphrase generation model. In training our sentence compression model, we investigate the impact of using different early-stopping criteria, such as embedding-based cosine similarity and F1. We utilize the pre-trained models (ours, GPT2 and T5) in different settings for single and multi-document summarization."]},{"key":"dc:title","label":"Title","values":["Abstractive multi-document summarization - paraphrasing and compressing with neural networks"]}]}],"canonical_facts":{"dc:date.issued":["2020"],"dc:description.other":["This thesis presents studies in neural text summarization for single and multiple documents.The focus is on using sentence paraphrasing and compression for generating fluent summaries, especially in multi-document summarization where there is data paucity. A novel solution is to use transfer-learning from downstream tasks with an abundance of data. For this purpose, we pre-train three models for each of extractive summarization, paraphrase generation and sentence compression. We find that summarization datasets – CNN/DM and NEWSROOM – contain a number of noisy samples. Hence, we present a method for automatically filtering out this noise. We combine the representational power of the GRU-RNN and TRANSFORMER encoders in our paraphrase generation model. In training our sentence compression model, we investigate the impact of using different early-stopping criteria, such as embedding-based cosine similarity and F1. We utilize the pre-trained models (ours, GPT2 and T5) in different settings for single and multi-document summarization."],"dc:identifier":["hdl:10133/5827"],"dc:subject":["Artificial intelligence","Automatic programming (Computer science)","Computer programming","Dissertations, Academic","Machine learning","Natural language processing (Computer science)","Neural networks (Computer science)"],"dc:title":["Abstractive multi-document summarization - paraphrasing and compressing with neural networks"],"dc:type":["Thesis"]},"updated_at":"2026-07-27T20:02:41Z"}