We propose a solution for detecting and summarizing human activities from a large amount of video. A video is divided into short segments(called stories), and a set of simple motion/image features are computed for each image frame. Prototypical elementary action patterns(called action events) is constructed by vector quantization on these features.A bipartite graph is constructed by taking stor...

In a human dialogue it is usually considered inappropriate if one conversant monopolizes the conversation. Similarly it can be inappropriate for a natural language database interface to respond with a lengthy list of data. A non-enumerative "summary" response is less verbose and often avoids misleading the user where an extensional response might. In this paper we investigate the problem of gen...

In this paper we present an approach to summarizing positive and negative opinions in blog threads. We first run a sentiment analysis system and consequently pass its output through a standard LSA-based text summarization system. Further on, we evaluate our approach and present the results obtained, which we believe are promising in the context of multi-document text summarization. Finally, we ...

Of course, setting up a summarization model that ensures smooth integration with human thinking is a relatively long process. Fig. 1 shows where the researchers are located in such a research strategy: An empirical model describing the intellectual strategies of human expert summarizers has been elaborated and published (most comprehensively in Endres-Niggemeyer 1998). The empirical model has b...

In this paper we describe research on summarizing conversations in the meetings and emails domains. We introduce a conversation summarization system that works in multiple domains utilizing general conversational features, and compare our results with domain-dependent systems for meeting and email data. We find that by treating meetings and emails as conversations with general conversational fe...

We define temporal summaries of news stories as extracting as few sentences as possible from each event within a news topic, where the stories are presented one at a time and sentences from a story must be ranked before the next story can be considered. We outline an evaluation strategy that we have developed for this task and describe simple language models for capturing novelty and usefulness...

This paper addresses the problem of summarizing decisions in spoken meetings: our goal is to produce a concise decision abstract for each meeting decision. We explore and compare token-level and dialogue act-level automatic summarization methods using both unsupervised and supervised learning frameworks. In the supervised summarization setting, and given true clusterings of decisionrelated utte...

This paper describes the multi-document summarization system designed by the Webclopedia team from ISI for DUC 2005. In contrast to the past DUCs and previous designs, this version of our summarizer consists of a query-interpretation component that directly analyzes the given user profile and topic narrative for each document cluster before creating a corresponding summary. This system ranks 4 ...