Syed Muhammad Hamza's Projects
An Android-based application build using the MVC model That allows users to Encipher(encrypt) and Decipher( decrypt) text using using following Ciphers - Shift Cipher - Vigenere Cipher - Substitution Cipher - Playfair Cipher
Awesome Knowledge Distillation
Classifying Consumer Finance Complaints into one of eleven product categories, The problem is a Text classification, also known as text tagging or text categorization. Text classifiers can automatically analyze text and then assign a set of pre-defined tags or categories based on its content. In this problem, I have taken 'consumer_complaint_narrative' as “text” and to classify each consumer_complaint_narrative / “text” into one of eleven pre-defined categories of product.
Portfolio for my Projects
Neural Network from scratch in Python to recognize handwritten digit achieved 98.45% test accuracy and using Keras CNN(Convolutional neural network) achieved 99.25% test accuracy deployed model to production
Regression-based Movie Recommender system that's a hybrid of content-based and collaborative filtering Recommender system Simply rate some movies and get immediate recommendations tailored for you
Tokenize and Parse a String of Characters to identify if the string is a valid java program Developed java lexical and syntactical analyzer from scratch using (Compilers: Principles, Techniques, and Tools) book as a reference,
A predictive analysis on california housing prices using machine learning supervised algorithms
The purpose of this repository is self-motivation and to keep track of my Machine learning, Natural Language Processing & Data Science related stuff progress
Myers–Briggs Type Indicator (MBTI) classification where my classifier can classify your personality type based on Isabel Briggs Myers self-study Myers–Briggs Type Indicator (MBTI). The classification result can be further used to match people with the most compatible personality types
It is the first Natural Language Processing competition on Kaggle I took part in Using logistic regression with test accuracy is 77.03% using Embedding and LSTM Recurrent Neural Network test accuracy is 98%
Collected images from google through web-scraping performed data cleaning, data preprocessing, exploratory data analysis, and build machine learning models such as Logistic Regression, Random Forest, and SVM(Support vector machine) achieved 88% test accuracy and deployed model to production, Used Numpy, OpenCV, SKlearn, CSS, Html, Flask, JavaScript, Selenium