This paper presents a video summarization technique for an Internet video to provide a quick way to overview its content. This is a challenging problem because finding important or informative parts of the original video requires to understand its content. Furthermore the content of Internet videos is very diverse, ranging from home videos to documentaries, which makes video summarization much ...

In this paper, we described our participation in the question-focused multi-document summarization task of DUC 2005. Our system was based on a supervised machine learning method in which feature extraction was an important issue. We present the whole procedure of our system, focusing on the features we used. We also analyze the results of manual and automatic evaluations.

Usually, presence of the same information in multiple documents is the main problem faced in effective information access. Instead of this redundant information thus accessed or retrieved, users are interested in retrieving information that addresses one or other several aspects. In such situation, text summarization proves to be very useful. Not only in Information retrieval, but it is an extr...

ROUGE stands for Recall-Oriented Understudy for Gisting Evaluation. It includes measures to automatically determine the quality of a summary by comparing it to other (ideal) summaries created by humans. The measures count the number of overlapping units such as n-gram, word sequences, and word pairs between the computer-generated summary to be evaluated and the ideal summaries created by humans...

We present results from using a version of the pyramid method to create gold standards for evaluation of automatic text summarization techniques in the domain of governmental texts. Our results show that the pyramid method may be useful to create gold standards for extraction based summarization techniques using only five human summarisers.

As enormous amount of electronic documents on the Web have been increasing, the necessity of automatic summarization has also been increasing to help people grasp the essential points of the documents. Many summarization techniques dealing with single document and multi-documents have been studied. However, due to the increase of the documents which report the change of topics along a timeline,...

In this paper I detail the implementation of an extractionbased summarization system that uses sentence clustering and named entity identification as main features for the 2006 Multilingual Summarization Evaluation. I discuss some of the failings of my system, and what can be done to improve it.

Most multi-document summarizers utilize term frequency related features to determine sentence importance. No empirical studies, however, have been carried out to isolate the contribution made by frequency information from that of other features. Here, we examine the impact of frequency on various aspects of summarization and the role of frequency in the design of a summarization system. We desc...

How to use provenance to explain why a query returns a result or why a result is missing has been studied extensively. Recently, we have demonstrated how to uniformly answer these types of provenance questions for first-order queries with negation and have presented an implementation of this approach in our PUG (Provenance Unification through Graphs) system. However, for realisticallysized data...