Installation

Install iscnlp-tagger, and then iscnlp-parser: Follow the instructions in the following links:

iscnlp / pos-tagger

iscnlp / parser

Clone the project:

git clone https://github.com/99percentcocoa/usr_python.git

cd usr_python

sudo pip install -r requirements.txt

sudo pip install .

Usage

Import the package in your project using

from usr_python import main

The USR function can then be called by main.getUSR(input), where input is the Hindi string whose USR is to be generated.

Output will be a python dictionary containing the sentence's USR representation.

---

Example output:

Input:

"सूर्यास्त के बाद आकाश को देखना कितना अच्छा लगता है।"

Output:

{
    "original": "# सूर्यास्त के बाद आकाश को देखना कितना अच्छा लगता है।",
    "concepts": [
        "sUryAswa_1",
        "AkASa_1",
        "xeKa_1-nA_cAhawA_hE_1",
        "kiwanA_1",
        "acCA+laga_1-wA_hE_1"
    ],
    "concepts_index": [
        "1",
        "2",
        "3",
        "4",
        "5"
    ],
    "semantic": [
        "",
        "",
        "",
        "",
        ""
    ],
    "gnp": [
        "",
        "[m sg a]",
        "[m sg -]",
        "",
        ""
    ],
    "dependency": [
        "0:k7t",
        "3:k2",
        "",
        "6:mod",
        ""
    ],
    "discourse": [
        "",
        "",
        "",
        "",
        ""
    ],
    "speaker_viewpoints": [
        "",
        "",
        "",
        "",
        ""
    ],
    "scope": [
        "",
        "",
        "",
        "",
        ""
    ],
    "sentence_type": "affirmative"
}