Image credit Research & Applied Projects
Projects that move from idea to algorithm, thinking with data and speaking through models.
Master's Project
Green Energy Demand Forecasting with Volatility-Aware LSTM
Developed a new UK daily dataset (2000–2023) and compared ARIMA, SARIMAX+GARCH, and deep learning models to forecast green energy demand.
Read MoreCourse Project
Interpretable Dialogue Generation with Attention-based Encoder-Decoder
Trained and interpreted an attention-based encoder–decoder model for dialogue generation, using beam search decoding and epoch-wise attention analysis.
Read MoreCourse Project
Context-Aware Dialogue Act Tagging with CNN-BiLSTM Architecture
Built a hybrid CNN-BiLSTM model for context-aware dialogue act tagging and visualized attention masks for interpretability.
Read MoreCourse Project
Character Similarity via Vector Space Semantics
Explored distributional semantics to measure character similarity in dialogue data using TF-IDF and cosine similarity.
Read MoreCourse Project
Image Classification on FashionMNIST with CNN
Built and trained a CNN on FashionMNIST using PyTorch with proper weight initialization, loss analysis, and performance evaluation.
Read MoreCourse Project
CNN Architecture Search with Genetic Algorithms
Used a Genetic Algorithm to evolve optimal CNN architectures for Farsi vowel classification, combining evolutionary computation with deep learning.
Read MoreCourse Project
London Tube Route Finder with Cost-Aware Search Algorithms
Developed a cost-aware path-finding system for London Underground using BFS, DFS, UCS, and zone-based heuristics.
Read MoreCourse Project
Empirical Demonstration of LLN and CLT on Pareto and Cauchy Distributions
Simulated and visualized how LLN and CLT behave under heavy-tailed distributions such as Pareto and Cauchy in Python, contrasted with light-tailed cases.
Read MoreCourse Project
Distribution Modeling on Financial Time Series with MLE
Applied Maximum Likelihood Estimation on Dow Jones daily returns to identify the best-fitting distribution and compared empirical and theoretical PDFs.
Read MoreCourse Project
Sentiment Analysis from Tweets using SVM
Built a sentiment analysis pipeline for tweets using SVM with TF-IDF features, cross-validation, and error analysis.
Read MoreGraduation Project
Reality Capture from Multiple Video Cameras
Developed a multi-camera reality capture system using YOLO, MIDAS, and photogrammetry techniques for 3D vehicle reconstruction in Intelligent Transportation Systems.
Read MoreInternship Project
Time Series Forecasting for Siemens Electrical Product Sales
Developed a SARIMA-based forecasting solution for Siemens Turkey to predict future sales of electrical products, supporting demand planning and inventory optimisation.
Read MoreCourse Project
Logic and Digital System Design Projects (FPGA Implementation)
Overview This project involved designing and implementing digital systems using Verilog on Xilinx Vivado, deployed to an Artix-7 FPGA board.
Read MoreCourse Project
Functional Enrichment Analysis of Differentially Expressed Genes (MCF7 – Estradiol)
Overview This project focused on identifying the biological pathways and cellular functions affected by estradiol treatment in MCF7 breast cancer cell lines.
Read MoreCourse Project
Gene Expression Clustering and Visualisation using Machine Learning
Overview This project explores how unsupervised machine learning methods can reveal hidden biological patterns in cancer-related gene expression data.
Read MoreCourse Project
Comparative Genomic Analysis of SARS-CoV-2
Overview This project focused on a comparative genomic and proteomic analysis of SARS-CoV-2, aiming to identify conserved genomic regions and mutations in the Spike protein that influence viral infectivity.
Read More