There are three different STMT files. Here are the details of which STMT files you should use:
File | Size | Contents | Features | Usages
|
---|
stmt${YEAR}api.jar | ~130 Kb | | - Find all sub-term related features without preload corpus
|
|
stmt${YEAR}dist.jar | ~12 Mb | - STMT classes
- HSqlDb classes
- lvg classes
- csv classes
| - Find all sub-term related features with preload corpus
- HSqlDb tables (not included)
- Normalization by applying lvg
|
|
stmt${YEAR}.tgz | ~900 Mb -> ~4.5 Gb | - stmt2013dist.jar
- HsqlDb data
- data files
- Java source codes
| - Find all sub-term related features with preload corpus
- HSqlDb tables
- Normalization by applying lvg
- For full installation
| - Java APIs
- STMT command line tools
|