tailieunhanh - Báo cáo khoa học: "An Automatic Method for Summary Evaluation Using Multiple Evaluation Results by a Manual Method"

To solve a problem of how to evaluate computer-produced summaries, a number of automatic and manual methods have been proposed. Manual methods evaluate summaries correctly, because humans evaluate them, but are costly. On the other hand, automatic methods, which use evaluation tools or programs, are low cost, although these methods cannot evaluate summaries as accurately as manual methods. In this paper, we investigate an automatic evaluation method that can reduce the errors of traditional automatic methods by using several evaluation results obtained manually. . | An Automatic Method for Summary Evaluation Using Multiple Evaluation Results by a Manual Method Hidetsugu Nanba Faculty of Information Sciences Hiroshima City University 3-4-1 Ozuka Hiroshima 731-3194 Japan nanba@ Manabu Okumura Precision and Intelligence Laboratory Tokyo Institute of Technology 4259 Nagatsuta Yokohama 226-8503 Japan oku@ Abstract To solve a problem of how to evaluate computer-produced summaries a number of automatic and manual methods have been proposed. Manual methods evaluate summaries correctly because humans evaluate them but are costly. On the other hand automatic methods which use evaluation tools or programs are low cost although these methods cannot evaluate summaries as accurately as manual methods. In this paper we investigate an automatic evaluation method that can reduce the errors of traditional automatic methods by using several evaluation results obtained manually. We conducted some experiments using the data of the Text Summarization Challenge 2 TSC-2 . A comparison with conventional automatic methods shows that our method outperforms other methods usually used. 1 Introduction Recently the evaluation of computer-produced summaries has become recognized as one of the problem areas that must be addressed in the field of automatic summarization. To solve this problem a number of automatic Donaway et al. 2000 Hirao et al. 2005 Lin et al. 20o3 Lin 2004 Hori et al. 2003 and manual methods Nenkova et al. 2004 Teufel et al. 2004 have been proposed. Manual methods evaluate summaries correctly because humans evaluate them but are costly. On the other hand automatic methods which use evaluation tools or programs are low cost although these methods cannot evaluate summaries as accurately as manual methods. In this paper we investigate an automatic method that can reduce the errors of traditional automatic methods by using several evaluation results obtained manually. Unlike other automatic methods our .