Dietary Meal Recommendation System using Neural NetworksEve Bytes(2024)
A dietary food recommendation system is a tool that is designed to help businesses in the food industry improve customer satisfaction and loyalty by providing personalized recommendations for food and drinks based on an individual's preferences and past orders. The primary objective of a food recommendation system is to increase sales and revenue by encouraging customers to try new menu items and by encouraging repeat business. This can be achieved by analyzing customer data and using machine learning algorithms to understand patterns in customer behavior and preferences, and by presenting recommendations that are tailored to the individual's tastes and needs. By using a food recommendation system, businesses can improve their customer retention rates, increase customer satisfaction, and ultimately drive growth and profitability. Dietary food recommendation systems also help individuals who are on a diet and trying to maintain a healthy lifestyle plan out their daily meal intake. In this project, we will develop, train and test a Machine Learning model that analyzes users' dietary preferences, health goals, and nutritional requirements, ultimately providing personalized meal plans and recommendations. The solution will be used in a website application or even a mobile app. The problem is an unsupervised, offline and a model-based learning problem, since we are building a recommendation model from the whole unlabled data.
The scope of this project involved creating a Machine Learning model which recommends dietary food products to consumers based on attributes such as nutritional information and recipes. The model could then be integrated into a mobile or web application using a backend infrastructure.
Handmade Product Recommendation System using NLP and Collaborative FilteringEve Bytes (2024)
In this project, we will be integrating artificial intelligence into our platform to enhance user experience and engagement. The process will involve developing AI-powered features to personalize recommendations, improve search functionality, and streamline user interactions.
The scope of this project involved creating a Machine Learning model which recommends handmade products to consumers based on attributes such as product description and user reviews. The model could then be integrated into a mobile or web application using a backend infrastructure.
The model was developed using the Python programming language with the Jupyter Notebook environment. The main purpose of using Jupyter Notebook instead of an IDE such as PyCharm or Sublime Text is because Jupyter Notebook allows you to integrate markdown text along with Python code within one file. The Python libraries involved in the development process include Pandas, NumPy, Scikit-Learn, Seaborn, and Matplotlib.
Bone Fracture Classification using Mura DatasetEziline Software House (Pvt). Ltd (2023)
Here we attempt to create an algorithm to classify images for bone fracture, for this an input image is given and output is given as a “Positive” or “Negative” label. The input data is in the form of X Ray images of the bones. Hence, an appropriate supervised learning model is to be trained with the data to give correct label to the input image to predict a fracture. Some preprocessing of the data (converting RGB images to Grayscale) is also necessary here. This repository contains a Keras implementation of a 169 layer Densenet Model on MURA dataset
Email Spam Filtering using Text ClassificationEziline Software House (Pvt). Ltd (2023)
Email becomes a powerful tool for communication as it saves a lot of time and cost. It is one of the most popular and secure medium for online transferring and communication messages or data through the web. But, due to the social networks, most of the emails contain unwanted information which is called spam. To identify such spam email is one of the important challenges. In this project we will use PYTHON text classification technique to identify or classify email spam message. We will find accuracy, time and error rate by applying suitable algorithms (such as NaiveBayes, NaiveBayesMultinomial and J48 etc.) on Email Dataset and we will also compare which algorithm is best for text classification.
AI Content Generation using Python and OpenAIEziline Software House (Pvt). Ltd (2023)
Created an AI based content generation application for creating and generating content on various topics. The tools used were:
- Python
- Flask
- OpenAI
- VS Code
- React JS
Carshop Application using Java and IntelliJ IdeaUniversity of York - Yorkshire, England (2023)
The carshop application is a console based app developed in Java and IntelliJ Idea which allows users to enter the information such as ID, manufacturer, model, year, mileage, size and price of a certain number of cars. The user can then perform the following tasks:
- Sort the cars based on the model
- Sort the cars based on the price
- Search the car with the lowest mileage
- Search the car with the lowest price
- Search the car based on the ID entered
Price Prediction using Linear RegressionDice Analytics (2022)
This is a tickets pricing monitoring system. It scrapes tickets pricing data periodically and stores it in a database. Ticket pricing changes based on demand and time, and there can be significant difference in price. We are creating this product mainly with ourselves in mind. Users can set up alarms using an email, choosing an origin and destination (cities), time (date and hour range picker) choosing a price reduction over mean price.
Portfolio Website using GitShowCaseDouglas College - Vancouver, Canada (2022)
In this project, I created a portfolio on GitShowCase showcasing different projects related to Web Development. The tools and technologies used for this project are:
- VS Code
- Github
- HTML
- CSS
- JavaScript
Restaurant Menu using Java and EclipseDuke University Online (2020)
This project is a console based application that allows a user to browse through a restaurant menu. The user can select from a number of food options and calculate the total bill. The code of this project was written in Java using classes and OOP. The IDE used is Eclipse.
BS Thesis in SMS Spam DetectionUniversity of Lahore - Islamabad, Pakistan (2018 - 2019)
Completed Research and Thesis on SMS Spam Detection Using Machine Learning and Android Studio.
- Collected Raw Datasets and compiled them
- Developed AI model using Machine Learning, SVM and Random Forest
- Wrote code in Python and Java
- Developed an app in Android Studio
- Wrote an 80 page thesis on final year project