Decayed DivRank for Guided Summarization

نویسندگان

  • Pan Du
  • Jipeng Yuan
  • Xianghui Lin
  • Jin Zhang
  • Jiafeng Guo
  • Xueqi Cheng
چکیده

Guided summarization is essentially an aspect-based multi-document summarization, where aspects can be taken as specified queries in summarization. We proposed a novel ranking algorithm, Decayed DivRank (DDRank) for guided summarization tasks of TAC2011. DDRank can address relevance, importance, diversity, and novelty simultaneously through a decayed vertex-reinforced random walk process in sentence ranking. With sentence ranked by DDRank, top-ranked sentences are supposed to be able to present aspect-related, important, diverse, and novel information contained in documents. Hence, we can select the top-ranked sentences as summary candidates directly. Aspects are approximately obtained by pLSA, and are adopted to generate sentence prior for DDRank in our approach.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Supervised Aggregation Framework for Multi-Document Summarization

In most summarization approaches, sentence ranking plays a vital role. Most previous work explored different features and combined them into unified ranking methods. However, it would be imprecise to rank sentences from a single point of view because contributions from the features are onefold in these methods. In this paper, a novel supervised aggregation approach for summarization is proposed...

متن کامل

TMSP: Topic Guided Manifold Ranking with Sink Points for Guided Summarization

Guided summarization is an extension of query-focused multidocument summarization. We proposed a novel ranking algorithm, Topic Guided Manifold Ranking with Sink Points (TMSP) for guided summarization tasks of TAC2010. TMSP is a topic extended version of Manifold Ranking with Sink Points (MRSP), which handles the Update Summarization tasks of TAC2009 well. We adopt the TMSP and MRSP methods to ...

متن کامل

PolyU's Experimentation with Guided Summarization

In addition to the familiar tasks of update summarization and automatic summary evaluation, the TAC 2010 summarization track includes a new task: guided summarization. It is designed to “encourage a deeper linguistic (semantic) analysis of the source documents” in order to generate a summary guided by a given set of aspects. The new task is an upgrade of query-focused summarization in that an a...

متن کامل

UofL at TAC 2011 Guided Summarization Task

In this paper, we describe our guided summarization system that participated in the TAC 2011 competition. We submitted two runs for the guided summarization task by following a random walk paradigm. Two different approaches were applied for the update component to create two runs of our guided summarization system: 1) using ROUGE (RecallOriented Understudy for Gisting Evaluation), and 2) using ...

متن کامل

WHUSUM Participation at TAC 2011 Guided Summarization Track

In this report, we present details about the participation of WHUSUM in the guided summarization track at TAC 2011. Guided summarization task requires participants to produce short, coherent summaries of news articles with the guidance of predefined categories and aspects for each category. This year, we extended our query-focused update summarization system with aspect related information. In ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2011