manasbichoo Goto Github PK

als

ALS algorithm recommender system

archbot

Chatbot made on python

blog-web_app-flask

Blog Web App using Python, Flask.

bookrecommenderknn

This is a book recommender system used to recommend top N books to the user using KNN Algorithm.

code_snippets

collegeproject

Initial under dev clg project

computer-vision

Computer Vision

content-engine

A very simple content-based recommendation engine. Great for learning, but also ready for real-world use.

covid-19-detection-using-chest-x-ray

Predict Covid-19 using chest x-ray scan

covid-racing-bar-chart

deeplearningprojects

All the deep learning projects

fastai

The fastai deep learning library, plus lessons and tutorials

flask-chat-app-article

flask-chatterbot

Simple boilerplate for ChatterBot using Flask

flask_pusher_chat

geeksforgeeksscrapper

Scrapes g4g and creates PDF

machine-learning-web-apps

Building and Embedding Machine Learning Model into a Web App(With Flask,etc)

manasbichoo.github.io

My Personal Website

mathematics_for_beginners

This is the formula sheet for "Mathematics for Beginners" by Siraj Raval on Youtube

mcq-web-app

A python Django MCQ Web App

menshut

news-analytics-nlp

nlp

NLP Projects

nlp-web-app-flask

Web Application to demonstrate NLP, made using Flask.

python

All Algorithms implemented in Python

python-1

:snake: Python Programs

recommendersystem

A recommender system on APACHE SPARK

rockpaperscissor-ai-

Building an AI to play Rock Paper Scissor using Neural Networks

songgenreclassifier

spellchecker

We used NLP(Natural Language Processing) to solve this problem. We used character embedding (chars2vec model) to generate vectorized words and plotted them on vector space to map the closest and semantically similar words together using cosine similarity as the metric. We used a pre trained NLP model called as chars2vec for this purpose. Having the character embedding, every single word’s vector can be formed even it is out-of-vocabulary words. This model is a CNN which performs the character embedding. The step by step solution for this problem statement is: Import the chars2vec model for character embedding. Import the text file and read the words stored in it. Make a array(list) of words of the file. Vectorize all the words of that file by chars2vec model’s vectorize method. Train the chars2vec model with training set as the list of vectorized words. Take an input word from the user. Vectorize the input word using chars2vec vectorize method. Find the cosine similarity of that input word with every other word from the text file’s list of words. The word with the highest cosine similarity is the output. Return the word with the highest cosine similarity which is the correct spelling of the misspelled word.