The use of MMR, diversity-based reranking for reordering documents and producing summaries

DMCA

The use of MMR, diversity-based reranking for reordering documents and producing summaries (1998)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Jaime Carbonell , Jade Goldstein

Venue:	In SIGIR
Citations:	765 - 14 self

BibTeX

@INPROCEEDINGS{Carbonell98theuse,
    author = {Jaime Carbonell and Jade Goldstein},
    title = {The use of MMR, diversity-based reranking for reordering documents and producing summaries},
    booktitle = {In SIGIR},
    year = {1998},
    pages = {335--336}
}

OpenURL

Abstract

jadeQcs.cmu.edu Abstract This paper presents a method for combining query-relevance with information-novelty in the context of text retrieval and summarization. The Maximal Marginal Relevance (MMR) criterion strives to reduce redundancy while maintaining query relevance in re-ranking retrieved documents and in selecting apprw priate passages for text summarization. Preliminary results indicate some benefits for MMR diversity ranking in document retrieval and in single document summarization. The latter are borne out by the recent results of the SUMMAC conference in the evaluation of summarization systems. However, the clearest advantage is demonstrated in constructing non-redundant multi-document summaries, where MMR results are clearly superior to non-MMR passage selection. 2 Maximal Marginal Relevance Most modem IR search engines produce a ranked list of retrieved documents ordered by declining relevance to the user’s query. In contrast, we motivated the need for “relevant novelty ” as a potentially superior criterion. A first approximation to measuring relevant novelty is to measure relevance and novelty independently and provide a linear combination as the metric. We call the linear combination “marginal relevance ”- i.e. a document has high marginal relevance if it is both relevant to the query and contains minimal similarity to previously selected documents. We strive to maximize-marginal relevance in retrieval and summarization, hence we label our method “maximal marginal relevanci ” (MMR).

Keyphrases

diversity-based reranking relevant novelty maximal marginal relevance retrieved document query relevance text summarization superior criterion user query ir search engine high marginal relevance non-redundant multi-document summary mmr diversity document retrieval summac conference minimal similarity text retrieval linear combination mmr result method maximal marginal relevanci maximize-marginal relevance single document summarization recent result non-mmr passage selection re-ranking retrieved document preliminary result ranked list first approximation summarization system apprw priate passage linear combination marginal relevance

DMCA