Some example projects that was made using Tensorflow (mostly). This repository contains the projects that I've experimented-tried when I was new in Deep Learning.

Persian word embedding ( نشاننده واژه ها فارسی | تعبیه سازی کلمات فارسی )

Word2Vec Library

Word2Vec Library

Make Models for The Quora Question Pairs competition was hosted on Kaggle to identify duplicate questions

Word2Vec Library

Word2Vec Library

build a word2vec model on hamshahri data and try to clustering embedded texts and find their real topic

Word2Vec Library

Word2Vec Library

Real-Time Solutions | Live Demo 1. Sentiment analysis of real-time tweets 2. Spam Ham Classifier

For every given pair of sentences -- (sentence-1, sentence-2), we need to determine if sentence-2 can be logically inferred given sentence-1.

Detection of misinformation of climate change using topic modeling (LDA) and Word Vectors

Literature Mining for Neurodegenerative disease

An evolving autoencoder using tensorflow. The idea is to have it evolve pictures, where fitness is closeness to some high dimensional point close to what we want. So like, if we know pictures of dogs exist as a point in some high dimensional encoded space, if we can evolve a picture to have en encoding close to the dogs, maybe we can get a dog l…

This script is written for AI-powered matchmaking system that intelligently pairs users based on their textual responses. It uses Word2Vec embeddings to understand the semantic meaning of responses and cosine similarity to determine how closely two users' responses align.

BOW,Word-to-vec to do sentiment classification into positive or negative on movie dataset

exploring machine learning methodologies for citation link prediction within academic networks. This project leverages Natural Language Processing, graph theory, and advanced feature engineering to identify potential citation links, demonstrating an evolutive development process from baseline approaches to optimized, performance-driven solutions.

Analyzing the tweets by the people and then correctly predicting their sentiments