# HMM-NER

**Repository Path**: jostree/HMM-NER

## Basic Information

- **Project Name**: HMM-NER
- **Description**: Use a hmm model to recognize the named entity in Chinese.
- **Primary Language**: C++
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2015-05-14
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

#HMM-NER

##build method
1. Make sure that there is cmake in your computer. If not, you can use `sudo apt-get install cmake`.
   to install cmake in ubuntu or `brew install cmake` to install cmake in Mac OS.
2. Create a build folder in the project dir. You can us this command `mkdir build`.
3. Go in the folder `build` with command `cd build`, and input the command `cmake ..` to generate the `makefile`.
4. Type the command `make` to build all the project.

##Usage
1. There are five executable files in the folder `build/bin`:
    * `trainhmmwithtag` can load People's Daily corpus, output the transmit probability and emit
        probability. *USAGE: * `./trainhmmwithtag input output`.
    * `gentestset` can generate the test set and the right tag answer from Pelple's Daily corpus.
        **USAGE: ** `./gentestset input output`.
    * `vit` is the decoder can recongise a word is a entity or not and give the tag.**USAGE**: `./vit
        model <input >output`.
    *`evaluate` is used to evaluate the tag generated form `vit` is effective or not. 
    **USAGE: **`./evaluate yourtagedfile answer`.