Skip to content

Latest commit

 

History

History
36 lines (24 loc) · 2.39 KB

README.md

File metadata and controls

36 lines (24 loc) · 2.39 KB
Raw

Mango Guru AI (NLP Chatbot)

License Follow

A simple NLP AI mango bot built in python built for the Mango website.

This repository contains a mango chatbot built using a combination of natural language processing (NLP) techniques and machine learning. The mangobot is trained on a dataset of pre-defined intents and their corresponding patterns and tags. The chatbot is able to classify the user's input into one of the pre-defined intents, and respond with a pre-defined message corresponding to the classified intent.

Dependencies

  • nltk (Natural Language Toolkit): This module is used for a variety of natural language processing tasks, including tokenization (breaking up a string into individual words) and lemmatization (reducing a word to its base form).
  • numpy: This module is used to store and manipulate the training data in the form of arrays.
  • tensorflow: This module is used to build and train the chatbot's neural network model. It provides tools for defining the model architecture, compiling the model, and fitting the model to the training data.

Usage

To use train and run the mangobot by yourself, run the following command:

model, words, classes = initialize()

This will initialize the chatbot by loading the dataset, preprocessing it, and training a neural network model on it.

To classify the user's input and generate a response, run the following command:

classify(model, words, classes, user_input)

Replace user_input with the desired user input. This will return the classified intent and corresponding response.

Data

The chatbot is trained on the intents.json file, which contains a list of pre-defined intents. Each intent consists of a list of patterns and a tag. The patterns are used to train the chatbot to classify user inputs into the corresponding intent, while the tag is used to determine the response to generate for each intent.

Model

The chatbot uses a neural network model with three dense layers and dropout layers in between to classify user inputs into the pre-defined intents. The model is trained using the Adam optimization algorithm and categorical cross-entropy loss. The accuracy of the model is also tracked during training.