Voting Algorithm for DoS attack prediction combining BERT for log analysis and Random Forest for IP packet analysis
Cyber Security, BERT, LLG, ML, linux system logs, IP packets, Random Forest Classifier,
More Info on projectAnalyzing data streams to prevent DNS attacks
Cyber Security, ML, DL, Kafka, Dockers
More Info StreamsComparison of cGANs vs CycleGANs for Image Augmentation by Sketching
cGAN, CycleGAN, Data Augmentation
More Info GANsWaiting Time predictions in amusement parks using Data Science Machine / Deep Learning
ML DL
More Info projectStoryPoints and time Estimation on Technology Projects using NLP and BERT
NLP, BERT, Agile, UserStories, StoryPoints
More Info NLPImproving Lottery Ticket Hypothesis with Knowledge Distillation
DL LTH Knowledge Distaillation NMIST
More Info LTH/KDThis project aims to learn how to host a model online using Google Cloud. Platform. This is part of a bigger project in which I extract data from a web page and use it to calculate how long it would take for a visitor to ride a rollercoaster in an amusement park. In this specific case, I give a full tutorial on GCP on how to create the necessary elements to host a pre-trained machine learning model and query it from anywhere around the world.
This was one of my favourite projects. The main objective is to predict, based on specific features , how long it would take an average park visitor to ride their favourite rollercoaster. This is a complete project as it involves data scrapping, data analysis and data science, a deep understanding of classical machine learning algorithms, and the understanding of deep learning algorithms such as Multilayer Perceptrons and Recurrent Neural Networks, to name a few examples. You can see my complete tutorials and results, together with all the code I did, below.