https://github.com/akullpp/bhmm
A Bayesian Hidden Markov Model Tagger
https://github.com/akullpp/bhmm
Last synced: 4 days ago
JSON representation
A Bayesian Hidden Markov Model Tagger
- Host: GitHub
- URL: https://github.com/akullpp/bhmm
- Owner: akullpp
- License: mit
- Created: 2014-03-23T12:21:16.000Z (over 12 years ago)
- Default Branch: master
- Last Pushed: 2017-11-03T11:58:03.000Z (over 8 years ago)
- Last Synced: 2025-01-16T16:42:50.435Z (over 1 year ago)
- Language: Java
- Size: 73.2 KB
- Stars: 4
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# Bayesian Hidden Markov Model
## Usage
```
mvn clean install
java -jar target/BHMM-1.0-SNAPSHOT.jar
```
## Arguments
```
[FINE | FINER | FINEST] detail of logging.
FINE = Basic information
FINER = Information about variables
FINEST = Information about variables in each sampling step
```
## Configuration
Via `configuration.properties`:
* corpus: Corpus file, see example corpus.txt
* lexicon: Lexicon file, see example lexicon.txt
* gold: Gold standard file, see example gold.txt
* out: Output file
* alpha: Hyperparameter for transitions
* beta: Hyperparameter for emissions
* iterations: Number of sampling iterations
* max: Maximum temperature
* min: Minimum temperature
* decrease: Iteration steps at which the temperature is decreased by the rate
* rate: Rate of decrease
* dbg: Iteration steps at which the FINEST information is written in the log file
## License
MIT