Data

Tools

Indexes

Applications

Experiments

Open Source Science

persian-postagger-with-lstm

A Persian POS Tagger with LSTM

https://github.com/minasmz/persian-postagger-with-lstm

Science Score: 26.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (5.9%) to scientific vocabulary

Keywords

lstm nlp persian-language persian-nlp persian-pos-tagger persian-postagger pos-tagger
Last synced: 9 months ago · JSON representation

Repository

A Persian POS Tagger with LSTM

Basic Info
  • Host: GitHub
  • Owner: minasmz
  • Language: Python
  • Default Branch: master
  • Homepage:
  • Size: 80.7 MB
Statistics
  • Stars: 13
  • Watchers: 0
  • Forks: 5
  • Open Issues: 0
  • Releases: 1
Topics
lstm nlp persian-language persian-nlp persian-pos-tagger persian-postagger pos-tagger
Created over 8 years ago · Last pushed almost 4 years ago
Metadata Files
Readme Citation

README.md

Persian-POSTagger-with-LSTM

A Persian POS Tagger with LSTM

In this Part of Speech Tagger I used an Implementation from here which has used LSTM neural network. you can look at the pdf file in this repository and you can watch this video for more information. I have trained the model with Hamshahri corpus and have tested on it, and get the glove file from here to train.

This model reaches 94.7 percent accuracy (tested on hamshahri testing corpus).

In second commit I added makeCompatible.py which normalizes persian texts. I have used this code, and slightly have changed it for tokenizing and normalizing input text.

For running this POS Tagger put your arbitrary input text in user.txt file, and call model_evaluation.py and get word/POS-tag format of your input.

You can see a sample output of this LSTM POS Tagger below:

/N /ADJ /N /N /N /ADJ /N /P /N /N /N /DELM /N /N /N /N /ADJ /CON /NUM /P /N /N /P /N /V /P /N /N /ADJ /V ./DELM /N /N /P /N /ADJ /V /CON /ADV /CON /N /N /ADJ /N /N /DELM /N /CLITIC /N /V ./DELM /N /N /N /DELM /P /N /N /N /V /CON /ADV /P /DET /N /N /CON /N /V ./DELM /N /P /N /CON /N /N /N /ADJ /ADJ /V /CON /P /DET /ADJ /P /N /CON /N /DELM /N /DELM /N /DELM /N /DELM /N /DELM /N /DELM /N /DELM /N /N /DELM /N /DELM /N /DELM /N /DELM /N /DELM /N /DELM /N /CON /N /ADJ /ADJ /V /CON /N /DET /N /CLITIC /ADJ /V ./DELM

GitHub Events

Total
  • Watch event: 1
Last Year
  • Watch event: 1