1 / 29

A progressive sentence selection strategy for document summarization

Presenter : Bo- Sheng Wang Authors : You Quyang , Wenjie Li, Renxian Zhang, Qin Lu IPM, 2013. A progressive sentence selection strategy for document summarization. Outlines. Motivation Objectives Methodology Experiments Conclusions Comments. Motivation.

yakov
Download Presentation

A progressive sentence selection strategy for document summarization

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Presenter : Bo-Sheng Wang Authors : You Quyang, Wenjie Li, Renxian Zhang, Qin Lu IPM, 2013 A progressive sentence selection strategy for document summarization

  2. Outlines Motivation Objectives Methodology Experiments Conclusions Comments

  3. Motivation Since there are actually many overlapping concepts in the input documents, it is indeed unnecessary and redundant to repeatedly mention one concept in the summary.

  4. Objectives They mainly consider the problem of how to construct summaries with good saliency and coverage.

  5. Methodology In this paper, they propose a novel sentence selection strategy that follows a progressive way to select the summary sentences

  6. Methodology Step 1: Define the subsuming relationship between two sentences.

  7. Methodology Step 1: Define the subsuming relationship between two sentences.

  8. Methodology Step 1: Define the subsuming relationship between two sentences. The relationship between two sentences is determined by the relations between the concepts.

  9. Methodology Step 1-1: The targetis to study the subsuming relations between the words in the input documents. Linguistic relation database (WordNet) Frequency-based statistics (co-occurrence)

  10. Methodology Step 1-1: The targetis to study the subsuming relations between the words in the input documents.

  11. Methodology They expect the relations to have the characteristics listed below. Sentence-level coverage. Set-based coverage Transitive reduction

  12. Methodology-Sentence-level coverage • In document summarization, sometimes a document set just consists of only a few documents. (For example:10 documents per set in the DUC 2004 data.) • They intend to study the sentence-level co-occurrence statistics instead of document-level co-occurrence.

  13. Methodology-Set-based coverage • Sentence-level co-occurrence is sparser than document-level co-occurrence due to the shorterlength of sentences. • Therefore ,the sentence-level coverage of a word with respect to another is usually much smaller. • They intend to examine the coverage not only between two words, but also between a word and a word set. (For example:there are two common phrases ‘‘King Norodom’’ and‘‘Prince Norodom’’. In the input documents, the coverage of ‘‘Norodom’’ with respect to either ‘‘King’’ or ‘‘Prince’’ isnot large enough and thus ‘‘Norodom’’ is not recognized to be subsumed by any one of the two. On the other side,‘‘Norodom’’ is almost entirely covered by the set {‘‘King’’, ‘‘Prince’’}. Therefore, if we can define a set-based coverage,more relations can be discovered)

  14. Methodology-Transitive reduction They also conduct a transitive reduction on the relations. i.e: to three words a, b, c that satisfy a > b, b > c and a > c (a > b denotes a subsuming b), the long-term relationship a > c will be ignored, since we prefer to include the subsuming word b into the summary before including the subsumed word c.

  15. Methodology-Necessary measures • Spanned Sentence Set(SPAN): • SPAN(w): The Spanned Sentence Set of a word win document set D • SD:Sentence Set • SPAN(w)={s|sϵSD ^ wϵs}= Define as the set of the sentences. • Concept Coverage(COV): • COV(w|W)=|SPAN(w)∩∪iSPAN(wi)|/|SPAN(W)| =Defined as the proportion of the sentences in SPAN(w) that appear in SPAN(W).

  16. Methodology

  17. Methodology Step1-2 : (1)They define the concept of “Connected Word” i.e: W={w1,…..,wl}; W’={w’1,…..,w’m} condition: w l1; . . . ; wlkϵW∪W, s.t.wi < wl1 ^ w11 < wl2 ^ . . . ^ wl(k-1) < wlk ^ wlk< w’1 (2)The Conditional Saliency of calculated as a weighted sum of the importance of all the ‘‘connected words’’ CS(s|s’)ΣwiϵsLOG(MAXw’jϵs’CON(wi|w’j * score(wi)))

  18. Methodology Step 2:

  19. Methodology • Step : • Toevery word that is not subsumed by any other word, we regard it as a general word and attach it to ROOT-W. • we calculate the score of each unselectedsentence based on its conditional saliency to each selected sentence. • Formula: Score(s|Sold)=Max stϵ Sold{CS(s,st)} * 1/len(s) * (1-pos(s)) • Penalizing: Score(wi)=α * Score(wi)

  20. Experiments • Step: • Evaluated on a generic multi-document summarization data set. • Evaluated on a query-focused multi-document summarization data set. • Pre-processed: • Removing the stop-words and stemming the remaining words

  21. Experiments-Evaluation metrics • ROUGE • State-of-the-art automatic summarization evaluation • They mainly makes use of N-gram comparison. • DUC

  22. Experiments-Generic summarization

  23. Experiments-Generic summarization

  24. Experiments-Generic summarization

  25. Experiments-Generic summarization

  26. Experiments-Query-focused summarization

  27. Conclusions Progressive system consistently performs better than the sequential system on every data set. The method competes comparably with the best submitted systems. The results clearly demonstrate the advantages of the progressive sentence selection strategy in constructing summaries with better saliency and coverage.

  28. Comments • Advantages • The method that have better saliency and coverage. • In unsupervised case, find the number of categories can be save some time. • Applications • Object Discovery

  29. Comments • Advantages • The method that have better saliency and coverage. • Disadvantage • The method spend some time than traditional methods. • Applications • Sentence selection

More Related