Multi-document Summarization Research Articles

Effective Elytron Vespid-B rank BiLSTM classifier for Multi-Document Summarization

Effective Elytron Vespid-B rank BiLSTM classifier for Multi-Document Summarization

A novel multi document summarization with document-elements augmentation for learning materials using concept based ILP and clustering methods

Multi Document Summarization (MDS) is a technique for extracting succinct summaries from groups of related documents. The usage of MDS in the e-learning context is more appealing for providing summaries for learning materials, which helps students and teachers to focus on key concepts of the learning materials. In a learning material, availability of non-textual document-elements such as figures/diagrams, plots, graphs, tables and algorithms can be seen widely, but keeping such document-elements in the summary is not possible as they are non-textual. This proposed work incorporates the text summarization approach with document-elements augmentation to the summary to provide a detailed coverage of information without exceeding the summary length constraint. The key information in the source text is identified by important phrase features and sentence features, and the summaries are generated by selecting important sentences using the Integer Linear Programming (ILP) framework while reducing the redundancy using pre-trained sentence vectors. The relationships between the summary and document-elements are identified through document-element snippet extraction and a Hierarchical Agglomerative Clustering approach. Experimental results of the proposed summary extraction and augmentation on educational dataset (EduSumm) show better performance compared to the state-of-the-art approaches.

Decomposition–based multi-objective differential evolution for extractive multi-document automatic text summarization

The central challenge in Automatic Text Summarization (ATS) involves efficiently generating machine-generated text summaries through optimization algorithms. An ATS is a critical component for systems dealing with textual information processing. However, the current approach faces a significant hurdle due to the computational intensity of the process, particularly when employing complex optimization techniques like swarm intelligence optimization alongside a costly ATS repair operator. While the current approach yields impressive Recall-Oriented Understudy for Gisting Evaluation (ROUGE) metrics for the generated summary, it struggles with inefficiencies, mainly attributed to the substantial optimization time consumed by the ATS repair operator scheme. In order to address this, a novel solution called Decomposition-based Multi-Objective Differential Evolution (MODE/D) is proposed. It is built upon the foundation of Differential Evolution for Multi-Objective Optimization (DEMO) and the weighted sum method (WS), coupled with an innovative ATS repair operator scheme. Through experimentation on Document Understanding Conferences (DUC) datasets, the novel approach of MODE/D – WS is validated by evaluating the results using ROUGE metrics. The outcomes are twofold: a remarkable reduction in serial execution time and a noteworthy enhancement over existing techniques in the scholarly domain, as evidenced by improved ROUGE-1, ROUGE-2, and ROUGE-L scores.

Review on Query-focused Multi-document Summarization (QMDS) with Comparative Analysis

The problem of query-focused multi-document summarization (QMDS) is to generate a summary from multiple source documents on identical/similar topics based on the query submitted by the users. This article provides a systematic review of the literature of QMDS. The research works are classified into six major categories based on the summarization methodologies used. Different techniques used for finding query-relevant summaries for different algorithms under each of the six major groups are reported. Further, 17 evaluation metrics used for evaluating algorithms for text summaries against the human-curated summaries are compiled here in this article. Extensive experiments are performed on eight different datasets. Comparative results of nine methodologies, each representing one of the six different groups, are presented. Seven different evaluation metrics are used in the comparative study. It is observed that DL- and ML-based QMDS methods perform. better in comparison to the other methods.

A Multi Document Summarization of Learning Materials using Bigram Embedding Technique and Integer Linear Programming

In the present era of the Internet, teachers and learners are heavily inclined to use e-learning systems for an efficient learning process. Due to the proliferation of educational text contents in these e-learning systems, the need for incorporating advanced text analysis tools and techniques are becoming inevitable. Multi Document Summarization (MDS) is a technique for producing concise summaries from a collection of related text documents. The usage of MDS in the context of e-learning is more promising for providing summaries for learning materials which helps students and teachers to focus on key concepts of the learning materials. In this paper a semantic approach towards the learning material summarization is proposed based on bigram embedding and ILP technique. This approach considers bigram as the basic meaningful semantic unit of the sentences to understand and summarize documents. Embedding techniques are employed to learn the vector representation of phrases to semantically identify similar phrases to reduce the redundancy and improving coherence. Using ILP technique, the summaries were generated by selecting important sentences while reducing the redundancy using phrase vectors. Experimental results on newly created educational dataset (EduSumm) shows better performance compared to the baseline systems.

Occams: A Text Summarization Package

Extractive text summarization selects asmall subset of sentences from a document, which gives good “coverage” of a document. When given a set of term weights indicating the importance of the terms, the concept of coverage may be formalized into a combinatorial optimization problem known as the budgeted maximum coverage problem. Extractive methods in this class are known to beamong the best of classic extractive summarization systems. This paper gives a synopsis of thesoftware package occams, which is a multilingual extractive single and multi-document summarization package based on an algorithm giving an optimal approximation to the budgeted maximum coverage problem. The occams package is written in Python and provides an easy-to-use modular interface, allowing it to work in conjunction with popular Python NLP packages, such as nltk, stanza or spacy.

Compressed Heterogeneous Graph for Abstractive Multi-Document Summarization

Multi-document summarization (MDS) aims to generate a summary for a number of related documents. We propose HGSum — an MDS model that extends an encoder-decoder architecture to incorporate a heterogeneous graph to represent different semantic units (e.g., words and sentences) of the documents. This contrasts with existing MDS models which do not consider different edge types of graphs and as such do not capture the diversity of relationships in the documents. To preserve only key information and relationships of the documents in the heterogeneous graph, HGSum uses graph pooling to compress the input graph. And to guide HGSum to learn the compression, we introduce an additional objective that maximizes the similarity between the compressed graph and the graph constructed from the ground-truth summary during training. HGSum is trained end-to-end with the graph similarity and standard cross-entropy objectives. Experimental results over Multi-News, WCEP-100, and Arxiv show that HGSum outperforms state-of-the-art MDS models. The code for our model and experiments is available at: https://github.com/oaimli/HGSum.

SumREN: Summarizing Reported Speech about Events in News

A primary objective of news articles is to establish the factual record for an event, frequently achieved by conveying both the details of the specified event (i.e., the 5 Ws; Who, What, Where, When and Why regarding the event) and how people reacted to it (i.e., reported statements). However, existing work on news summarization almost exclusively focuses on the event details. In this work, we propose the novel task of summarizing the reactions of different speakers, as expressed by their reported statements, to a given event. To this end, we create a new multi-document summarization benchmark, SumREN, comprising 745 summaries of reported statements from various public figures obtained from 633 news articles discussing 132 events. We propose an automatic silver-training data generation approach for our task, which helps smaller models like BART achieve GPT-3 level performance on this task. Finally, we introduce a pipeline-based framework for summarizing reported speech, which we empirically show to generate summaries that are more abstractive and factual than baseline query-focused summarization approaches.

Optimizing Multidocument Summarization by Blending Reinforcement Learning Policies

We consider extractive summarization within a cluster of related texts (multi-document summarization). Unlike single-document summarization, redundancy is particularly important because sentences across related documents might convey overlapping information. Thus, sentence extraction in such a setting is difficult because one will need to determine which pieces of information are relevant while avoiding unnecessary repetitiveness. To solve this difficult problem, we propose a novel reinforcement learning based method <inline-formula><tex-math notation="LaTeX">$\mathbf{PoBRL}$</tex-math></inline-formula> (<b>Po</b>licy <b>B</b>lending with maximal marginal relevance and <b>R</b>einforcement <b>L</b>earning) for solving multi-document summarization. PoBRL jointly optimizes over the following objectives necessary for a high-quality summary: importance, relevance, and length. Our strategy decouples this multi-objective optimization into different sub-problems that can be solved individually by reinforcement learning. Utilizing PoBRL, we then blend each learned policies to produce a summary that is a concise and a complete representation of the original input. Our empirical analysis shows high performance on several multi-document datasets. Human evaluation also shows that our method produces high-quality output.

Plan and generate: Explicit and implicit variational augmentation for multi-document summarization of scientific articles

Multi-Document Summarization of Scientific articles (MDSS) is a challenging task that aims to generate concise and informative summaries for multiple scientific articles on a particular topic. However, despite recent advances in abstractive models for MDSS, grammatical correctness and contextual coherence remain challenging issues. In this paper, we introduce EDITSum, a novel abstractive MDSS model that leverages sentence-level planning to guide summary generation. Our model incorporates neural topic model information as explicit guidance and sequential latent variables information as implicit guidance under a variational framework. We propose a hierarchical decoding strategy that generates the sentence-level planning by a sentence decoder and then generates the final summary conditioned on the planning by a word decoder. Experimental results show that our model outperforms previous state-of-the-art models by a significant margin on ROUGE-1 and ROUGE-L metrics. Ablation studies demonstrate the effectiveness of the individual modules proposed in our model, and human evaluations provide strong evidence that our model generates more coherent and error-free summaries. Our work highlights the importance of high-level planning in addressing intra-sentence errors and inter-sentence incoherence issues in MDSS.

Grapharizer: A Graph-Based Technique for Extractive Multi-Document Summarization

In the age of big data, there is increasing growth of data on the Internet. It becomes frustrating for users to locate the desired data. Therefore, text summarization emerges as a solution to this problem. It summarizes and presents the users with the gist of the provided documents. However, summarizer systems face challenges, such as poor grammaticality, missing important information, and redundancy, particularly in multi-document summarization. This study involves the development of a graph-based extractive generic MDS technique, named Grapharizer (GRAPH-based summARIZER), focusing on resolving these challenges. Grapharizer addresses the grammaticality problems of the summary using lemmatization during pre-processing. Furthermore, synonym mapping, multi-word expression mapping, and anaphora and cataphora resolution, contribute positively to improving the grammaticality of the generated summary. Challenges, such as redundancy and proper coverage of all topics, are dealt with to achieve informativity and representativeness. Grapharizer is a novel approach which can also be used in combination with different machine learning models. The system was tested on DUC 2004 and Recent News Article datasets against various state-of-the-art techniques. Use of Grapharizer with machine learning increased accuracy by up to 23.05% compared with different baseline techniques on ROUGE scores. Expert evaluation of the proposed system indicated the accuracy to be more than 55%.

Multi-Document News Web Page Summarization Using Content Extraction and Lexical Chain Based Key Phrase Extraction

In the area of text summarization, there have been significant advances recently. In the meantime, the current trend in text summarization is focused more on news summarization. Therefore, developing a synthesis approach capable of extracting, comparing, and ranking sentences is vital to create a summary of various news articles in the context of erroneous online data. It is necessary, however, for the news summarization system to be able to deal with multi-document summaries due to content redundancy. This paper presents a method for summarizing multi-document news web pages based on similarity models and sentence ranking, where relevant sentences are extracted from the original article. English-language articles are collected from five news websites that cover the same topic and event. According to our experimental results, our approach provides better results than other recent methods for summarizing news.

Discharge summary hospital course summarisation of in patient Electronic Health Record text with clinical concept guided deep pre-trained Transformer models

Brief Hospital Course (BHC) summaries are succinct summaries of an entire hospital encounter, embedded within discharge summaries, written by senior clinicians responsible for the overall care of a patient. Methods to automatically produce summaries from inpatient documentation would be invaluable in reducing clinician manual burden of summarising documents under high time-pressure to admit and discharge patients. Automatically producing these summaries from the inpatient course, is a complex, multi-document summarisation task, as source notes are written from various perspectives (e.g. nursing, doctor, radiology), during the course of the hospitalisation. We demonstrate a range of methods for BHC summarisation demonstrating the performance of deep learning summarisation models across extractive and abstractive summarisation scenarios. We also test a novel ensemble extractive and abstractive summarisation model that incorporates a medical concept ontology (SNOMED) as a clinical guidance signal and shows superior performance in 2 real-world clinical data sets.

Result diversification with negative type distances by multi-objective evolutionary algorithms

The result diversification problem is to select an optimal subset with high “quality” and “diversity” from a given ground set of items, which is popular in various applications such as web-based search, multi-document summarization and ensemble pruning. The diversity relies on the distance between items. Previous works mainly focused on metric distances, and applied a greedy or local search algorithm with theoretical guarantees. As a kind of global search algorithm inspired by Darwin’s theory of evolution, evolutionary algorithms (EAs) can have a better optimization ability than greedy and local search, but often lack theoretical support. Recently, EAs have been introduced to result diversification, achieving good theoretical guarantees besides superior empirical performances. In this paper, we study whether EAs can still achieve good theoretical guarantees for result diversification with negative type distances, which are also a class of important dissimilarity measures, especially in information retrieval and sketching techniques. We propose to reformulate the result diversification problem with negative type distances as a bi-objective maximization problem, and solve it by multi-objective evolutionary algorithms (MOEAs). We prove that a simple MOEA (i.e., GSEMO) can achieve the best-known polynomial-time approximation ratio. Experiments are also performed to examine the performance of different MOEAs on the application of web-based search.

BioASQ-QA: A manually curated corpus for Biomedical Question Answering

The BioASQ question answering (QA) benchmark dataset contains questions in English, along with golden standard (reference) answers and related material. The dataset has been designed to reflect real information needs of biomedical experts and is therefore more realistic and challenging than most existing datasets. Furthermore, unlike most previous QA benchmarks that contain only exact answers, the BioASQ-QA dataset also includes ideal answers (in effect summaries), which are particularly useful for research on multi-document summarization. The dataset combines structured and unstructured data. The materials linked with each question comprise documents and snippets, which are useful for Information Retrieval and Passage Retrieval experiments, as well as concepts that are useful in concept-to-text Natural Language Generation. Researchers working on paraphrasing and textual entailment can also measure the degree to which their methods improve the performance of biomedical QA systems. Last but not least, the dataset is continuously extended, as the BioASQ challenge is running and new data are generated.

Selection Informative Units for Extractive Summarization

An Extractive Multi-Document Summarizer must select the most informative units and prevents duplication in extraction. In order to achieve this goal, a new technique, called “comprising at least one Representative Term at the Highest Frequency”, called RTHF, is proposed in this work. The units which include representative terms, but with low frequencies are not considered for extraction (selection of the most informative units). On the other hand, these units which provide RTHF feature, precede other similar units in ranking (prevents duplication). The heuristic behind the RTHF is explained by probability. RTHF was experimented on a previously developed and tested paragraph- based Extractive Multi-Document Summarizer. The results show that it enhances the original system by 0.8% ~ 3.2% (Average-F values of ROUGE metrics).

Automatic multi-documents text summarization by a large-scale sparse multi-objective optimization algorithm

AbstractDue to the exponential overflow of textual information in various fields of knowledge and on the internet, it is very challenging to extract important information or to generate a summary from some multi-document collection in a specific field. With such a gigantic amount of textual content, human text summarization becomes impractical since it is expensive and consumes a lot of time and effort. So, developing automatic text summarization (ATS) systems is becoming increasingly essential. ATS approaches are either extractive or abstractive. The extractive approach is simpler and faster than the abstractive approach. This work proposes an extractive ATS system that aims to extract a small subset of sentences from a large multi-document text. First, the whole text is preprocessed by applying some natural language processing techniques such as sentences segmentation, words tokenization, removal of stop-words, and stemming to provide a structured representation of the original document collection. Based on this structured representation, the ATS problem is formulated as a multi-objective optimization (MOO) problem that optimizes the extracted summary to maintain the coverage of the main text content while avoiding redundant information. Secondly, an evolutionary sparse multi-objective algorithm is developed to solve the formulated large-scale MOO. The output of this algorithm is a set of non-dominated summaries (Pareto front). A novel criterion is proposed to select the target summary from the Pareto front. The proposed ATS system has been examined using (DUC) datasets, and the output summaries have been evaluated using (ROUGE) metrics and compared with the literature.

SMATS: Single and Multi Automatic Text Summarization

In today’s world, the rapid growth of textual data on internet sites & online resources makes it challenging for human being to assimilate essential information. To handle such issues, text summarization (TS) plays an important role. Through the TS process, a shorter version of the original content is generated to preserve the relevant information. This study suggests a quantitative assessment of models for single and multi-document summarization based on the sentence scoring method. Experimentation of the models has been carried out on DUC datasets. A detailed comparative analysis of the models is reported with respect to the performance of algorithms based on various metrics such as Recall Oriented-Understudy for Gisting Evaluation (ROUGE), Range, Co-efficient of Variation (CV) and Readability score.

Multi-Document Summarization Using Selective Attention Span and Reinforcement Learning

Multi-Document Summarization Using Selective Attention Span and Reinforcement Learning

Personalized Multi-document Text Summarization using Deep Learning Techniques

The summary of a large piece of text or a document is a concise description of the same thing. It must retain most of the important points from the document and remove any kind of verbosity. The text summarization task is an information processing scheme that given a document or a set of documents, mines the most essential content from the source by considering the user or task at hand and presents the summary in a well-formed and brief text. Summarization can be done not just on one document, it can be done on a set of documents as well, which is called multi-document summarization. The proposed approach considers and in-corporates user preferences for the summarization task. Depending on the current user or current task, the summary could differ essentially. For an instance, a summary on ‘photosynthesis’, would be very different for the kid in 4th Grade vs. a student doing a bachelor's in Science. Therefore, while writing a summary, the reading history of the user is employed as a source of personalizing text summarization tasks. The proposed method is experimentally evaluated in the domain of news articles and obtained better summaries capable of extracting important concepts based on user preferences explained in the document when considering the relevant domain terms in the process of multi-document text summarization.