nluanalizer 1.0.2

Natural Language Tool to Extract Features from Intents Dataset.

Introduction

This package will help the user to extract ngrams, nouns, verbs, adverbs and more features from an Intent Dataset in Spanish. This will help the user to balance the dataset by comparing number of ocurrences of each feature in each intent.

The entry format is an excel (.xlsx). Each column would be an intent and in the first row it has to be informed the intent name, then in the next rows the sentences for each intent training.

The output format is a json. It will have one dictionary per intent including all the features extracted.

How to use it?

Installation

pip install nluanalyzer

Usage

Usage

nluanalyzer -h|--help
nluanalyzer -v|--version
nluanalyzer <excel_path>

Options

-h --help                    Show help screen.
-v --version                 Show version.

Example

$ nlu_analyzer  intents.xlsx
'Successfully saved intents analyzed file in /home/project/location/Formatted_Metrics.xlsx'

Package Requirements

• nltk==3.4.5

• spacy==2.2.4

• docopt==0.6.2

• numpy==1.17.4

• docopt==0.6.2

• pandas==0.25.3

• textacy==0.10.0

• scikit_learn==0.23.1

History

1.0.0 (2020-05-26)

• Spanish Analysis Implemented.