You are here

New release of BFSU PowerConc August 2016


A new release of BFSU PowerConc (7 August 2016) is available now. This release is the 'beta 25b' of BFSU PowerConc Version 1.0, if you want to distinguish it from the more popular 'beta 21c' release.

                         PowerConc logo
                           Click and download!

What's new in beta 25b?

Some revisions were made for this release. For example,

1) The default concordance span is now set to 12 words to the left and 12 words to the right. The context display can be further extended to 50 words for either side of the context from the Concordance Settings.

2) In previous releases, concordance results stop at sentence boundaries. This limit is suppressed in the new release to allow for more context for KWIC type of concordance interpretation.

3) The word count of Chinese texts was refined to leave out some punctuations (e.g. douhao and Chinese dash) in the word list. Some optimisations were also made to minimise the possibility of miscalculating Chinese characters and words.

4) Negative keywords are moved to the end of keyword list and the minus sign added to negative keyword log-likelihood ratio values. Relative frequencies of observed corpus and reference corpus are shown in the result as well.

5) A checkbox of 'Collocation & Colligation' module was renamed, i.e. Use Outer Wordlist -->Use Ready Wordlist.

6) In the 'texts' folder, 15 sample English texts (taken from the Crown corpus representing the 15 text types of Brown corpus taxonomy) and 15 tokenised/segmented Chinese texts (taken from ToRCH2009) are enclosed as demo data.

7) More than a dozen configuration or resource files are included in the 'sys' folder, such as the lists of 10,000 Chinese idioms, color terms and shell nouns for batch search, CLAWS PoS tagset reduction (i.e. 137 tags to ten-somthing broad PoS categories), search result regrouping configuration files (e.g. 15 Brown corpus categories to four broad genre types). We can upload your files here, if would like to share with us your useful configuration and resource files created for your own research.

NB: New functionalities of corpus analysis will be mainly implemented in our next generation CorpusCloud system which to a great extent retains the BFSU PowerConc features.

Previous releases

BFSU PowerConc Version 1.0 beta 21c
BFSU PowerConc Version 1.0 beta 20

Bug reports can be sent to

How to cite BFSU PowerConc?

  1. Xu, Jiajin & Jia Yunlong. (2013). The design and development of the R-gram based corpus analysis tool ‘PowerConc’. Computer-Assisted Foreign Language Education (1): 57-62.
  2. Xu, Jiajin & Yunlong Jia. (2013). PowerConc: An R-gram based corpus analysis tool. Paper presented at American Association of Corpus Linguistics 2013 (AACL2013) Conference. San Diego, CA., 18-20 January.


BFSU PowerConc 2016年8月7日新出1.0 beta 25c版




  2)此前版本的PowerConc索引行显示时,只显示句内语境,不显示句末标点以外的文字内容。1.0 beta 25c版,为便于用户更全面地观察语境,取消了这一设置。


  4)主题词分析功能作了修改。上一版对正负主题词不作区分。1.0 beta 25c版将主题词置于主题词表底部,并在代表主题性的对数似然率值前添加负号。在结果显示中,观察语料库和参照语料库的相对频率(百分比)也一并给出。

  5)Collocation & Colligation功能模块中的界面文字Use Outer Wordlist改为了Use Ready Wordlist。


  7)在sys文件夹中,我们提供了10多个资源列表和配置文件。比如,我们准备了可用户批量检索的1万条常用汉语成语、英语颜色词表、英语外壳名词,以及对分析结果进行分组呈现的分组配置文件(测试用的15个文件可被分为Press、General、Learned和Fiction四类)。CLAWS赋码集中的137类词性码可以通过Reduce POS配置文件归并成10类左右的常用词类。用户也可将个人使用中编制的资源列表和配置文件发到,我们将上传到这里供更多用户使用。

  注:今后BFSU PowerConc将不会进行大的升级。有关功能方面的提升,将主要通过语料云在线分析系统实现。




  BFSU PowerConc Version 1.0 beta 21c
  BFSU PowerConc Version 1.0 beta 20






Resource packs:

  1. BNC word list to be used as the reference list in keyword generation.
  2. COCA word list to be used as the reference list in keyword generation.

(C) 北京外国语大学中国外语与教育研究中心

Belongs to: