素还真x素续缘漫画:How to use GoTagger

来源:百度文库 编辑:中财网 时间:2024/05/04 16:44:59
Top > Corpus Linguistics Softwares > How to use GoTagger


GoTagger Version 0.7 ... download (400KB)

GoTagger is a GUI-based Part-Of-Speech (POS) tagger that is freely availabe for research and education. This software is written in Delphi and thus runs on Windows wihout relying on any ActiveX or DLLs. GoTagger annotates a text with POS information utilizing the rule files contained in Eric Brill‘s POS tagger. If you don‘t have it, please download at Eric Brill‘s website.


< System Requirements >

  • Windows 98/ME/2000/XP
  • Intel Pentium M Processor with 1.2GHz (or equivalent)
  • 256Mb of RAM (512Mb is recommended)
  • 20MB of free disk space
  • Super VGA (800 x 600) or higher-resolution video adapter and monitor

< How to use >

GoTagger can be installed through the following steps.
  1. Download "GoTagger.zip" and unzip it into a folder of your choice (e.g. "C:\GoTagger\").
  2. Download Brill‘s tagger if you haven‘t had yet.
  3. Copy the 10 rule files in "Bin_and_Data" folder in Brill‘s tagger, and paste them into the "G_data" folder in GoTagger as shown in the screenshots below.


Here is the main screen of GoTagger.


(1) Directory explorer (2) File explorer
You can select one or more files using the directory explore (1) and the file explore (2).
Double-clicking a file in (2) will put it into the right frame (5) of the main window.

(3)
Add ... The files highlighted in (2) will be added to (5). Add all ... All of the files listed in (2) will be added to (5). Remove ... The files highlighted in (5) will be removed. Remove all ... All of the files listed in (5) will be removed.
(4) START
Tagging will begin just after pressing this button.

(5) Selected File(s)
This frame shows the files that will be processed.

(6) Settings
Lexicon Choose one of the Lexicon files. Contextual Rule Choose one of the Contextual Rule files. Separator Choose your preferred separator. Destination of output files If "..\(original file)\Tagged\" is selected, the "Tagged" folder will be automatically created under the same folder as the original files. In this option, the output files will be saved there. If you are inclined to "Specify" the save folder, press the "locate" button to select a directory.
NOTICE -- Any of the old files having the same name of newly created files will be automaticaly overwritten. Tokenizer Check the box written "On" if you need to tokenize sentences before tagging them. Lemmatizer Check the box written "On" if you need to lemmatize words. To enable this function, you need to download "e_lemma.txt", complied by Prof. Yasumasa Someya, and put it into "G_data". (7) Preview

(8) Processing Time

(9) Status

When the tagging process has finished, the results will be automatically displayed as shown below.


(10) List of output files
The tagged files will be shown here.

(11) Preview
Clicking a file in (10) will show the preview of it here.

(12) Tag
The tagset used in GoTagger (and Brill Tagger) is displayed.

(13) Tab
You can change the screen focus between "Select Files" and "Result".


< UnInstall >
Just delete all the files in "GoTagger" folder.


Mail
Please feel free to send comments or suggestions for amendments and inprovement. Thank you.