Sub-Term Mapping Tools

Normalization Tool (nt)

- Introduction
- Set Up
- Test Run
- Input
- Output
- Options

Introduction

Nt (Normalization Tool) is used in STMT to normalize terms before mapping. Three normalization (applies Lexical Tools) are provided:

• LexNorm
• SynonymNorm
• LvgNorm

Set Up

Follow the installation instructions to install and run the nt program. Check on the following items only if you don't use the provided script to install Sub-Term Mapping Tools.

• CLASSPATH:
  1. include the STMT distribution jar file, ${STMT_DIR}/lib/stmt${YEAR}dist.jar, in your CLASSPATH.
  2. include the stmt top directory in your CLASSPATH.

• Configuration File: assign the full path of the top directory of stmt${YEAR} to a variable named ROOT_DIR in the configuration file, data/Config/stmt.properties.

• Lvg Configuration File: set the lvg configuration file (LVG_CONFIG_FILE) to the default (data/Config/lvg.properties) or your installed lvg directory in the stmt configuration file.

Test Run

• Run java program

  Enter the command:

  > nt -p
  - Please input a term (type "Ctl-d" to quit) >
  Saw Film
  --- LexItemNorm ---
  saw film
  
  --- SynonymNorm ---
  see film
  saw film
  
  --- LvgNorm ---
  film see
  film saw
  

  where:

  • nt: nt script to run Normalization Tool Java class
  • -p: set nt system option to show prompt (try -h option!)

Input

Free text (a word or term)

Output

Normalization results of the input term

Options

Please refer to design document