論文投稿: 會議還是期刊?

有了研究成果,接下來就是投稿將其發表,但是投稿到會議好還是期刊好?哪一個比較重要?哪一個影響力比較大?

曾經當過 IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI) 期刊主編 (Editor-in-Chief) 的 University of Notre Dame 資訊系的 Prof. Kevin Bowyer,於 2012 年寫過一篇文章來討論這件事,文章標題是「Mentoring Advice on “Conferences Versus Journals” for CSE Faculty」,CSE 就是資訊工程領域,文章頗有參考性,摘錄如下。

1. 作者引用一篇發表在 Communications of the ACM 的文章 (Freyne 2010),文章中以資訊領域裡的人工智慧 (Artificial Intelligence, AI) 領域為例,分析了 8,764 篇會議及期刊論文,以 Google Scholar Impact Factor 為指標,發現:

“The results highlight how leading conferences compare favorably to mid-ranking journals, surpassing the impact of journals in the bottom half of the traditional ISI Web of Knowledge ranking.”

也就是說,以 impact factor 來說,頂尖期刊還是比頂尖會議的等級再高上一些。

該文章也提到,以一個會議的 accept/reject rate 來代表一個會議的好壞(accept rate 低代表會議較好)雖已漸成常態,但分析這些會議的 accept/reject rate 和 impact factor 之間的相關性,發現事情可能沒那麼簡單:

“The view that conference rejection rates are a good proxy for conference did not hold up to scrutiny, reflecting a low coefficient of correlation between the rejection rate of conferences and their Google Scholar scores.”

有辦過會議的人應該了解,一個會議的 accept/reject rate 取決於許多因素,而這些因素不見得跟會議的好壞有關。

2. 作者引用另一篇發表在 Communications of the ACM 的文章 (Franceschet 2010),發現「最多產的作者」、「被引用次數最多的作者」、和「最有名望的作者(Turing Award winner)」之間的有趣關係:

“The concepts of productivity, popularity, and prestige are well separate in computer science… That is, the authors who write the most papers do not overlap with the authors whose papers get cited most, or with the researchers who win the Turing Award.”

分析 papers 數量和被引用的次數,發現「被引用次數最多的作者」、也就是所謂 high-impact scholars,比較常在期刊發表:

“High-impact scholars publish significantly less than prolific ones, and more frequently in journals.”

這篇文章總結,發表在好的期刊雖然不容易但有它的價值:

“The fundamental message for the computer science community is: although it is more difficult to get published in journals, the effort is ultimately rewarded with a higher impact. From a bibliometric perspective, the best strategy to gain impact seems to be that of publishing few, final, and well-polished contributions in archival journals instead of many premature “publishing quarks” in conference proceedings.”

3. 另一篇發表在 Scientometrics 的文章 (Eckmann 2012) 分析了三個頂尖的電腦視覺會議 (CVPR, ICCV, ECCV) 和三個頂尖的電腦視覺期刊 (IEEE Transactions on PAMI, IJCV, CVIU) 之間的關係,發現:

“About 30% of journal papers were based on conference papers by the same authors… Journal papers with priors are cited more than journal papers without priors… Conference papers that are priors are cited more than top-3 conference papers that are not priors.”

如果只寫一篇論文,寫 PAMI 期刊論文當然比寫會議論文好,但是如果二者都寫—先有一篇好的會議論文,再延伸成為 PAMI 論文—影響力可能更大。

4. 作者提到,有些人會拿會議的 accept/reject rate 來跟期刊的 accept/reject rate 來比,如果會議的 accept rate 較低,就認為會議比期刊好。事實上,由於會議和期刊審查過程的不同,如果要比,可能要拿會議的 accept rate 跟期刊的「直接 accept、不需要下一輪 revise」的 rate 比,而這比例以作者擔任 IEEE Transactions on PAMI 主編期間的經驗,大約是 100 篇裡面只有 1 篇的數量級,如果這麼比,會議的 accept rate 就不見得比較低了。

“One big difference here is that a journal submission may be revised in a substantial way and re-reviewed to determine if the revisions make the appropriate improvements. The timeline of conferences simply does not allow this. If a submission to a highly competitive conference is reviewed and found to need substantial revisions that should be checked by reviewers, then the submission ends up in the reject pile.”

5. 作者引用一篇發表在 Communications of the ACM 的文章 (Vardi 2009),提到會議審查不若期刊嚴謹:

“The reviewing process performed by program committees is done under extreme time and workload pressures, and it does not rise to the level of careful refereeing.”

另一篇發表在 Communications of the ACM 的文章 (Fortnow 2009) 也有類似的觀察:

“As our research areas continue to become more specialized a few to none of the PC members can properly judge the importance of most results.”

這與作者自身經驗相符,會議稿量太大,內容涵蓋太廣,許多審稿員必須要審查自己不那麼熟悉的論文。這篇文章還提到,許多好的會議,可能只篩選出 incremental papers,因為這些 papers 在已知的知識範圍內,比較安全,比較容易引起共鳴,比較容易得到審稿員「一致」的肯定:

“The most difficult decisions are made by consensus. This leads to an emphasis on safe papers (incremental and technical) versus those that explore new models and research direction outside the established core areas of the conference.”

此外,會議的截稿日期,也可能讓作者急就章送出論文。

6. Fortnow 2009 建議,如果會議的功能不做發表,而是回到它最基本的功能:

“By de-emphasizing their publication role, conferences can once again play their most important role: Bringing the community together.”

IEEE Signal Processing Society 鼓勵作者投稿到 IEEE Signal Processing Letters,若被接受,論文作者可以選擇是否到下一個 ICASSP 或 ICIP 會議發表他們的「期刊」論文。這是會議和期刊間建立聯結的例子,也可以避免重覆投稿的情況。

7. 作者總結,對於「會議還是期刊?」這個問題:”Don’t think either-or, think both-and.” 在最好的會議發表,也在最好的期刊發表,因為從以上討論可見,不論會議或期刊都有它的價值,建議 conference-only 或 journal-only 可能都是偏頗的。

This entry was posted in 科學研究 and tagged . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s