With vast amounts of text being available in electronic format, such as news and social media, automatic multi-document summarization can help extract the most important information. We present and evaluate a novel method for automatic extractive multi-document summarization. The method is purely combinatorial, based on bicliques in the bipartite word-sentence occurrence graph. It is particular...

Common ethanol detection methods are not applicable to cell culture media and microdialysates due to interference with medium constituents including amino acids and pH indicators. We present a novel GC-MS method for the accurate and precise analysis of ethanol in cell cultures and microdialysates. The method is based on the carbonate-catalyzed extractive pentafluorobenzoylation of ethanol and d...

In this paper we present a novel resampling model for extractive meeting summarization. With resampling based on the output of a baseline classifier, our method outperforms previous research in the field. Further, we compare an existing resampling technique with our model. We report on an extensive series of experiments on a large meeting corpus which leads to classification improvement in weig...

One of the main problems in research on automatic summarization is the inaccurate semantic interpretation of the source. Using specific domain knowledge can considerably alleviate the problem. In this paper, we introduce an ontology-based extractive method for summarization. It is based on mapping the text to concepts and representing the document and its sentences as graphs. We have applied ou...

The most successful approaches to extractive text summarization seek to maximize bigram coverage subject to a budget constraint. In this work, we propose instead to maximize semantic volume. We embed each sentence in a semantic space and construct a summary by choosing a subset of sentences whose convex hull maximizes volume in that space. We provide a greedy algorithm based on the GramSchmidt ...

We present a new supervised framework that learns to estimate automatic Pyramid scores and uses them for optimizationbased extractive multi-document summarization. For learning automatic Pyramid scores, we developed a method for automatic training data generation which is based on a genetic algorithm using automatic Pyramid as the fitness function. Our experimental evaluation shows that our new...

We introduce a novel, fully unsupervised method to extract keywords from meeting speech in real-time. Our approach represents text as a word co-occurrence network and leverages the k-core graph decomposition algorithm and properties of submodular functions. We outperform multiple baselines in a real-time scenario emulated from the AMI and ICSI meeting corpora. Evaluation is conducted against bo...

Water markets are developing as part of a Council of Australian Governments initiative to promote an efficient use of Australia's water resources. The consequences of these policies on river health is yet to be fully understood, but recognised as having significant interrelationships which need to be explored. This paper examines the consequences of introducing trade and allocating water for en...

In this paper, we present a supervised learning approach to training submodular scoring functions for extractive multi-document summarization. By taking a structured predicition approach, we provide a large-margin method that directly optimizes a convex relaxation of the desired performance measure. The learning method applies to all submodular summarization methods, and we demonstrate its effe...