Multi-document Summarization Research Articles

Ranking with Recursive Neural Networks and Its Application to Multi-Document Summarization

We develop a Ranking framework upon Recursive Neural Networks (R2N2) to rank sentences for multi-document summarization. It formulates the sentence ranking task as a hierarchical regression process, which simultaneously measures the salience of a sentence and its constituents (e.g., phrases) in the parsing tree. This enables us to draw on word-level to sentence-level supervisions derived from reference summaries.In addition, recursive neural networks are used to automatically learn ranking features over the tree, with hand-crafted feature vectors of words as inputs. Hierarchical regressions are then conducted with learned features concatenating raw features.Ranking scores of sentences and words are utilized to effectively select informative and non-redundant sentences to generate summaries.Experiments on the DUC 2001, 2002 and 2004 multi-document summarization datasets show that R2N2 outperforms state-of-the-art extractive summarization approaches.

Fast and Accurate Prediction of Sentence Specificity

Recent studies have demonstrated that specificity is an important characterization of texts potentially beneficial for a range of applications such as multi-document news summarization and analysis of science journalism. The feasibility of automatically predicting sentence specificity from a rich set of features has also been confirmed in prior work. In this paper we present a practical system for predicting sentence specificity which exploits only features that require minimum processing and is trained in a semi-supervised manner. Our system outperforms the state-of-the-art method for predicting sentence specificity and does not require part of speech tagging or syntactic parsing as the prior methods did. With the tool that we developed --- Speciteller --- we study the role of specificity in sentence simplification. We show that specificity is a useful indicator for finding sentences that need to be simplified and a useful objective for simplification, descriptive of the differences between original and simplified sentences.

Multi-Document Summarization Based on Two-Level Sparse Representation Model

Multi-document summarization is of great value to many real world applications since it can help people get the main ideas within a short time.In this paper, we tackle the problem of extracting summary sentences from multi-document sets by applying sparse coding techniques and present a novel framework to this challenging problem. Based on the data reconstruction and sentence denoising assumption, we present a two-level sparse representation model to depict the process of multi-document summarization. Three requisite properties is proposed to form an ideal reconstructable summary: Coverage, Sparsity and Diversity. We then formalize the task of multi-document summarization as an optimization problem according to the above properties, and use simulated annealing algorithm to solve it.Extensive experiments on summarization benchmark data sets DUC2006 and DUC2007 show that our proposed model is effective and outperforms the state-of-the-art algorithms.

An EDU-Based Approach for Thai Multi-Document Summarization and Its Application

Due to lack of a word/phrase/sentence boundary, summarization of Thai multiple documents has several challenges in unit segmentation, unit selection, duplication elimination, and evaluation dataset construction. In this article, we introduce Thai Elementary Discourse Units (TEDUs) and their derivatives, called Combined TEDUs (CTEDUs), and then present our three-stage method of Thai multi-document summarization, that is, unit segmentation, unit-graph formulation, and unit selection and summary generation. To examine performance of our proposed method, a number of experiments are conducted using 50 sets of Thai news articles with their manually constructed reference summaries. Based on measures of ROUGE-1, ROUGE-2, and ROUGE-SU4, the experimental results show that: (1) the TEDU-based summarization outperforms paragraph-based summarization; (2) our proposed graph-based TEDU weighting with importance-based selection achieves the best performance; and (3) unit duplication consideration and weight recalculation help improve summary quality.

Multi-document update summarisation using co-related terms for scientific articles and news group

In the current trend, as the information on the internet is exploding exponentially in the course of time, and as nearly 90% of information is stored in the form of text, we need an effective and concise approach to help the user to locate the main points from the document corpus. Thus, there arises a need to provide a high quality summary in order to allow the user to quickly locate the desired information from the document clusters. Our proposed work deals with a new summarisation technique named as multi-document update summarisation based on co-related terms, considering terms and their related terms as concepts for identifying interrelated concepts from the documents of the dynamic clusters. Update summarisation is achieved by two methods: 1 static approach; 2 dynamic approach. We have also addressed the sentence reordering and redundancy elimination technique based on concepts to improve the efficiency of the summary. The proposed dynamic updation of summary is compared with the existing COBWEB update summarisation algorithm considering scientific articles and news tracks as dataset. From the analysis it is inferred that our proposed approach gives better results considering F-measure as the performance metric.

Vertex Cover Algorithm Based Multi-document Summarization Using Information Content of Sentences

In recent times, the requirement for the generation of multi-document summary has gained a lot of attention among researchers. Multi-document summarization systems focus on generating compressed form of the documents which maintains the pertinent features of the original documents. Mostly, text summarization techniques use the sentence extraction technique where the salient sentences in the multiple documents are selected and presented as a summary. In our proposed system, we have developed a sentence extraction based multi-document summarization system using the principle of vertex cover algorithm which automatically selects relevant sentences that cover the predominant concepts of the input documents. This frame work represents the documents as a weighted undirected graph with sentences as the vertices and the similarity between the sentences as the edge weight between the corresponding vertices in the graph. The experimental result on the DUC 2002 data sets demonstrates the effectiveness of the proposed method in document summarization

Topic-sensitive multi-document summarization algorithm

Latent Dirichlet Allocation (LDA) has been used to generate text corpora topics recently. However, not all the estimated topics are of equal importance or correspond to genuine themes of the domain. Some of the topics can be a collection of irrelevant words or represent insignificant themes. This paper proposed a topic-sensitive algorithm for multi-document summarization. This algorithm uses LDA model and weight linear combination strategy to identify significance topic which is used in sentence weight calculation. Each topic is measured by three different LDA criteria. Significance topic is evaluated by using weight linear combination to combine the multi-criteria. In addition to topic features, the proposed approach also considered some statistics features, such as term frequency, sentence position, sentence length, etc. It not only highlights the advantages of statistics features, but also cooperates with topic model. The experiments showed that the proposed algorithm achieves better performance than the other state-of-the-art algorithms on DUC2002 corpus.

The Benchmark of Paragraph and Sentence Extraction Summaries on Outlier Document Filtering Applied Multi-Document Summarizer

We studied outlier document filtering (ODF) for extractive sentence summarization. Our results are superior compared to the average of the participant systems’ using DUC 2006. Furthermore, we add extractive paragraph summarization to the same system. It is surprising that the results are nearly the same for ROUGE metrics. Although extractive paragraph summarization has a better performance for precision, extractive sentence summarization has a slightly better performance on the recall and F-Score which is the harmonic mean of recall and precision. The ODF is successful for both extractive sentence and paragraph summarization. The similarity metric (match percent) suggested in the article prevents the domination of longer sentences/paragraphs on shorter sentences/paragraphs in selection. As a result, the ODF provides the flexibility of paragraph extraction instead of sentence extraction for simplicity and readability and less work load. DOI: http://dx.doi.org/10.5755/j01.itc.43.4.7010

SRRank: Leveraging Semantic Roles for Extractive Multi-Document Summarization

Extractive multi-document summarization systems usually rank sentences in a document set with some ranking strategy and then select a few highly ranked sentences into the summary. One of the most popular ranking algorithms is the graph-based ranking algorithm. In this paper, we investigate making use of semantic role information to enhance the graph-based ranking algorithm for multi-document summarization. We first parse the sentences and obtain the semantic roles, and then propose a novel SRRank algorithm and two extensions to make better use of the semantic role information. Our proposed algorithms can simultaneously rank the sentences, semantic roles and words in a heterogeneous ranking process. Experimental results on two DUC datasets demonstrate that our proposed algorithms significantly outperform a few baselines, and the semantic role information is validated to be very helpful for multi-document summarization.

Information Ordering with an Event‐Enriched Vector Space Model for Multi‐Document News Summarization

Information ordering is a nontrivial task in multi‐document summarization (MDS), which typically relies on the traditional vector space model (VSM) notorious for semantic deficiency. In this article, we propose a novel event‐enriched VSM to alleviate the problem by building event semantics into sentence representations. The mediation of event information between sentence and term, especially in the news domain, has an intuitive appeal as well as technical advantage in common sentence‐level operations such as sentence similarity computation. Inspired by the block‐style writing by humans, we base the sentence ordering algorithm on sentence clustering. To accommodate the complexity introduced by event information, we adopt a soft‐to‐hard clustering strategy on the event and sentence levels, using expectation–maximization clustering and K‐means, respectively. For the purpose of cluster‐based sentence ordering, the event‐enriched VSM enables us to design an ordering algorithm to enhance event coherence computed between sentence and sentence–context pairs. Drawing on the findings of earlier research, we also incorporate topic continuity measures and time information into the scheme. We evaluate the performance of the model and its variants automatically and manually, with experimental results showing clear advantage of the event‐based model over baseline and non‐event‐based models in information ordering for multi‐document news summarization. We are confident that the event‐enriched VSM has even greater potential in summarization and beyond, which awaits further research. © 2014 Wiley Periodicals, Inc.

Exploring actor–object relationships for query-focused multi-document summarization

Most research on multi-document summarization explores methods that generate summaries based on queries regardless of the users' preferences. We note that, different users can generate somewhat different summaries on the basis of the same source data and query. This paper presents our study on how to exploit the information regards how users summarized their texts. Models of different users can be used either separately, or in an ensemble-like fashion. Machine learning methods are explored in the construction of the individual models. However, we explore yet another hypothesis. We believe that the sentences selected into the summary should be coherent and supplement each other in their meaning. One method to model this relationship between sentences is by detecting actor---object relationship (AOR). The sentences that satisfy this relationship have their importance value enhanced. This paper combines ensemble summarizing system and AOR to generate summaries. We have evaluated this method on DUC 2006 and DUC 2007 using ROUGE measure. Experimental results show the supervised method that exploits the ensemble summarizing system combined with AOR outperforms previous models when considering performance in query-based multi-document summarization tasks.

A novel contextual topic model for multi-document summarization

Information overload becomes a serious problem in the digital age. It negatively impacts understanding of useful information. How to alleviate this problem is the main concern of research on natural language processing, especially multi-document summarization. With the aim of seeking a new method to help justify the importance of similar sentences in multi-document summarizations, this study proposes a novel approach based on recent hierarchical Bayesian topic models. The proposed model incorporates the concepts of n-grams into hierarchically latent topics to capture the word dependencies that appear in the local context of a word. The quantitative and qualitative evaluation results show that this model has outperformed both hLDA and LDA in document modeling. In addition, the experimental results in practice demonstrate that our summarization system implementing this model can significantly improve the performance and make it comparable to the state-of-the-art summarization systems.

Minimum redundancy and maximum relevance for single and multi-document Arabic text summarization

Automatic text summarization aims to produce summaries for one or more texts using machine techniques. In this paper, we propose a novel statistical summarization system for Arabic texts. Our system uses a clustering algorithm and an adapted discriminant analysis method: mRMR (minimum redundancy and maximum relevance) to score terms. Through mRMR analysis, terms are ranked according to their discriminant and coverage power. Second, we propose a novel sentence extraction algorithm which selects sentences with top ranked terms and maximum diversity. Our system uses minimal language-dependant processing: sentence splitting, tokenization and root extraction. Experimental results on EASC and TAC 2011 MultiLingual datasets showed that our proposed approach is competitive to the state of the art systems.

Trends in Multi-document Summarization System Methods

formation is knowledge if it is rightly applied. Information are stored with different formats in databases but retrieving such from different documents has been a challenge. People want ready-made information for the purpose of decision making in minimal time and thereby crave for summary of information. Automatic summarization helps in mining data and delivering timely and cogent information to users. These systems attempt to address the issue of data mining using different summarization methods. This paper discusses existing methods and state of the art in automatic summarisation system from recent articles. Achievement and challenges involve are also discussed.

FoDoSu: Multi-document summarization exploiting semantic analysis based on social Folksonomy

Multi-document summarization techniques aim to reduce documents into a small set of words or paragraphs that convey the main meaning of the original document. Many approaches to multi-document summarization have used probability-based methods and machine learning techniques to simultaneously summarize multiple documents sharing a common topic. However, these techniques fail to semantically analyze proper nouns and newly-coined words because most depend on an out-of-date dictionary or thesaurus. To overcome these drawbacks, we propose a novel multi-document summarization system called FoDoSu, or Folksonomy-based Multi-Document Summarization, that employs the tag clusters used by Flickr, a Folksonomy system, for detecting key sentences from multiple documents. We first create a word frequency table for analyzing the semantics and contributions of words using the HITS algorithm. Then, by exploiting tag clusters, we analyze the semantic relationships between words in the word frequency table. Finally, we create a summary of multiple documents by analyzing the importance of each word and its semantic relatedness to others. Experimental results from the TAC 2008 and 2009 data sets demonstrate the improvement of our proposed framework over existing summarization systems.

Using non-lexical features for identifying factual and opinionative threads in online forums

Subjectivity analysis essentially deals with separating factual information and opinionative information. It has been actively used in various applications such as opinion mining of customer reviews in online review sites, improving answering of opinion questions in community question–answering (CQA) sites, multi-document summarization, etc. However, there has not been much focus on subjectivity analysis in the domain of online forums. Online forums contain huge amounts of user-generated data in the form of discussions between forum members on specific topics and are a valuable source of information. In this work, we perform subjectivity analysis of online forum threads. We model the task as a binary classification of threads in one of the two classes: subjective (seeking opinions, emotions, other private states) and non-subjective (seeking factual information). Unlike previous works on subjectivity analysis, we use several non-lexical thread-specific features for identifying subjectivity orientation of threads. We evaluate our methods by comparing them with several state-of-the-art subjectivity analysis techniques. Experimental results on two popular online forums demonstrate that our methods outperform strong baselines in most of the cases.

Event graphs for information retrieval and multi-document summarization

With the number of documents describing real-world events and event-oriented information needs rapidly growing on a daily basis, the need for efficient retrieval and concise presentation of event-related information is becoming apparent. Nonetheless, the majority of information retrieval and text summarization methods rely on shallow document representations that do not account for the semantics of events. In this article, we present event graphs, a novel event-based document representation model that filters and structures the information about events described in text. To construct the event graphs, we combine machine learning and rule-based models to extract sentence-level event mentions and determine the temporal relations between them. Building on event graphs, we present novel models for information retrieval and multi-document summarization. The information retrieval model measures the similarity between queries and documents by computing graph kernels over event graphs. The extractive multi-document summarization model selects sentences based on the relevance of the individual event mentions and the temporal structure of events. Experimental evaluation shows that our retrieval model significantly outperforms well-established retrieval models on event-oriented test collections, while the summarization model outperforms competitive models from shared multi-document summarization tasks.

A multi-document summarization system based on statistics and linguistic treatment

The massive quantity of data available today in the Internet has reached such a huge volume that it has become humanly unfeasible to efficiently sieve useful information from it. One solution to this problem is offered by using text summarization techniques. Text summarization, the process of automatically creating a shorter version of one or more text documents, is an important way of finding relevant information in large text libraries or in the Internet. This paper presents a multi-document summarization system that concisely extracts the main aspects of a set of documents, trying to avoid the typical problems of this type of summarization: information redundancy and diversity. Such a purpose is achieved through a new sentence clustering algorithm based on a graph model that makes use of statistic similarities and linguistic treatment. The DUC 2002 dataset was used to assess the performance of the proposed system, surpassing DUC competitors by a 50% margin of f-measure, in the best case.

Text summarization using Wikipedia

Automatic text summarization has been an active field of research for many years. Several approaches have been proposed, ranging from simple position and word-frequency methods, to learning and graph based algorithms. The advent of human-generated knowledge bases like Wikipedia offer a further possibility in text summarization – they can be used to understand the input text in terms of salient concepts from the knowledge base. In this paper, we study a novel approach that leverages Wikipedia in conjunction with graph-based ranking. Our approach is to first construct a bipartite sentence–concept graph, and then rank the input sentences using iterative updates on this graph. We consider several models for the bipartite graph, and derive convergence properties under each model. Then, we take up personalized and query-focused summarization, where the sentence ranks additionally depend on user interests and queries, respectively. Finally, we present a Wikipedia-based multi-document summarization algorithm. An important feature of the proposed algorithms is that they enable real-time incremental summarization – users can first view an initial summary, and then request additional content if interested. We evaluate the performance of our proposed summarizer using the ROUGE metric, and the results show that leveraging Wikipedia can significantly improve summary quality. We also present results from a user study, which suggests that using incremental summarization can help in better understanding news articles.

Efficient calculation of sentence semantic similarity: a proposed scheme based on machine learning approaches and NLP techniques

Sentence semantic similarity plays a crucial role in a variety of applications such as Machine Translation, Information Retrieval, Question Answering and Multi-document Summarization. Considering the variability of natural language expression, sentence semantic similarity detection is not a trivial task. This paper tries to make use of Natural Language Processing (NLP) as well as machine learning techniques in order to propose a scheme for sentence semantic similarity. In the first part of the proposed scheme, i.e., the NLP section, different sets of linguistic features including string-based, semantic-based, Named Entity-based and syntax-based features are extracted. In the second part, machine learning algorithms are used to construct classification models on the extracted set of features. Experimental results in the first part indicate that extracted features are valid for sentence semantic similarity. Moreover, by comparing the performance of different classification algorithms in the second part, KNN seems to be the most successful algorithm. Overall, experimental results indicate that the proposed approach can be used to improve the performance of sentence semantic similarity detection especially in terms of accuracy.