Text summarization model based on maximum coverage problem and its variant

Abstract

We discuss text summarization in terms of maximum coverage problem and its variant. We explore some decoding algorithms including the ones never used in this summarization formulation, such as a greedy algorithm with performance guarantee, a randomized algorithm, and a branch-andbound method. On the basis of the results of comparative experiments, we also augment the summarization model so that it takes into account the relevance to the document cluster. Through experiments, we showed that the augmented model is superior to the best-performing method of DUC'04 on ROUGE-1 without stopwords.

DESAMC+DocSum: Differential evolution with self-adaptive mutation and crossover parameters for multi-document summarization

“…In recent years, with the appearance of work [28], optimizationbased approaches have been intensively investigated in extractive document summarization [59,60,67,62,34,[2][3][4]. It directly connected with the character of the extractive summarization, i.e., identification of informative sentences in documents by the nature is an optimization problem.…”

“…It directly connected with the character of the extractive summarization, i.e., identification of informative sentences in documents by the nature is an optimization problem. Takamura and Okumura [59] represented text summarization as maximum coverage problem with knapsack constraint. McDonald [46] presented an approximate dynamic programming approach to maximize the MMR criteria.…”

Detecting Topics from Twitter Posts During TV Program Viewing

“…Filatova et al [2] were the first to formulize the summary of sentences as a maximum covering problem and they proposed the greedy algorithm. Takamura et al [11] ran detailed comparison experiments of algorithms proposed for applying the maximum covering problem to summarizing sentences. According to those experiments, the performance of the Filatova method is a little weak, but its processing time is the fastest.…”

“…(The method for setting c i is described below.) The above problem is called the knapsack-constrained maximum covering problem and it is known as a NP-hard problem [2], [11]. Thus, we use the greedy algorithm shown in Figure 3.…”

iWIN: A Summarizer System Based on a Semantic Analysis of Web Documents

“…For example, we are looking for a summary that contains the information content of the source documents as faithfully as possible but, at the same time, we would eliminate text redundancy. A summarization technique generally aims at generating a summary whose information set has the maximum number of relevant information in terms of semantic units with respect to the original documents, also noted in literature as the Maximum Coverage Problem [16].…”