My Bachelors Projects
Throughout my Bachelor’s degree, I have worked on various projects that integrate data analysis, machine learning, and data visualization. These projects allowed me to apply theoretical knowledge to real-world challenges, enhance my technical skills, and explore diverse fields, from environmental science to healthcare and network analysis. Each project provided valuable insights and equipped me with specialized skills in programming, statistical analysis, and visualization techniques essential for effective data-driven decision-making.
Factors Affecting Bird Extinction
2023
In this project, I and my classmates explored the environmental and biological factors that contribute to the extinction of bird species. By analyzing data on migratory patterns, population size, and habitat types, we built multiple linear regression models to predict extinction risks for various bird species. The project highlighted how migratory birds with smaller populations face a significantly higher risk of extinction compared to larger, non-migratory species.
In this project, I and my classmates explored the environmental and biological factors that contribute to the extinction of bird species. By analyzing data on migratory patterns, population size, and habitat types, we built multiple linear regression models to predict extinction risks for various bird species. The project highlighted how migratory birds with smaller populations face a significantly higher risk of extinction compared to larger, non-migratory species.
First Year of Bachelors
2022/2023
2022/2023
Predicting a skin lesion’s state
2023
In this project, our team developed a machine learning models to classify skin lesions, focusing on identifying types linked to skin cancer. Using the PAD-UFES-20 dataset, which includes clinical images and metadata of various skin lesions, we preprocessed images with Multi-Otsu thresholding and Snake Contour segmentation to isolate key features like color, asymmetry, and texture. We tested models including K-Nearest Neighbors, Nearest Centroid, Logistic Regression, and Random Forest. Through grid and random search, we refined model performance, using metrics like accuracy, F1-score, and AUC-ROC to evaluate robustness and reliability. The final model shows potential as a clinical diagnostic tool, achieving optimal accuracy with Random Forest after tuning.
In this project, our team developed a machine learning models to classify skin lesions, focusing on identifying types linked to skin cancer. Using the PAD-UFES-20 dataset, which includes clinical images and metadata of various skin lesions, we preprocessed images with Multi-Otsu thresholding and Snake Contour segmentation to isolate key features like color, asymmetry, and texture. We tested models including K-Nearest Neighbors, Nearest Centroid, Logistic Regression, and Random Forest. Through grid and random search, we refined model performance, using metrics like accuracy, F1-score, and AUC-ROC to evaluate robustness and reliability. The final model shows potential as a clinical diagnostic tool, achieving optimal accuracy with Random Forest after tuning.
Data Science in Research, Business and Society
First Year Project
Public Transport & CO2 Emissions
2022
This project explored how public transport systems in Denmark, Estonia, Romania, and Germany can be improved to boost usage and reduce CO2 emissions. Using interviews and online reviews, my classmates and I identified common issues such as delays, outdated infrastructure, and limited networks, along with positives like affordable fares. We proposed country-specific improvements, focusing on the need for electric transportation and better infrastructure to lower emissions and make public transport more appealing. This research demonstrates how effective public transport systems can help combat global warming by reducing CO2 output.
This project explored how public transport systems in Denmark, Estonia, Romania, and Germany can be improved to boost usage and reduce CO2 emissions. Using interviews and online reviews, my classmates and I identified common issues such as delays, outdated infrastructure, and limited networks, along with positives like affordable fares. We proposed country-specific improvements, focusing on the need for electric transportation and better infrastructure to lower emissions and make public transport more appealing. This research demonstrates how effective public transport systems can help combat global warming by reducing CO2 output.
- Data Modeling and Statistical Analysis
- Regression Analysis and ANOVA
- Outlier Management
- Data Visualization with ggplot2 in R
Skills Gained:
- Data Preprocessing & Image Segmentation
- Machine Learning Modeling (KNN, Logistic Regression, Nearest Centroid, Random Forest)
- Feature Extraction (Asymmetry & Color)
- Model Evaluation & Hyperparameter Tuning (Cross-validation, Accuracy, F1-score, and AUC-ROC)
- Critical Analysis & Research Writing
Skills Gained:
Skills Gained:
- Interviewing and Data Collection
- Qualitative Data Analysis
- Critical Thinking
- Research Writing
Clothing Classification Using Machine Learning Models
2023-2024
Сollaborating closely with my team, we explored various methods for classifying clothing types from images. We implemented Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and Naive Bayes classifiers to approach the task. Through extensive experimentation and analysis, we evaluated the accuracy of each model in recognizing clothing items based on visual features, optimizing the performance across different algorithms.
Сollaborating closely with my team, we explored various methods for classifying clothing types from images. We implemented Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and Naive Bayes classifiers to approach the task. Through extensive experimentation and analysis, we evaluated the accuracy of each model in recognizing clothing items based on visual features, optimizing the performance across different algorithms.
Second Year of Bachelors
2023/2024
2023/2024
Network Analysis
Machine Learning
Аnalysis of a Power Grid Network Robustness
2023-2024
My classmates and I utilized advanced analytical techniques to analyze and predict the network's resilience against targeted attacks, random failures, and cascading failures. Collaboratively, we conducted thorough analyses to assess the network's robustness under various scenarios, identifying potential vulnerabilities and points of failure. Additionally, we used Gephi for visualizing and analyzing the network.
My classmates and I utilized advanced analytical techniques to analyze and predict the network's resilience against targeted attacks, random failures, and cascading failures. Collaboratively, we conducted thorough analyses to assess the network's robustness under various scenarios, identifying potential vulnerabilities and points of failure. Additionally, we used Gephi for visualizing and analyzing the network.
- Machine Learning model development (CNN, SVM, Naive Bayes)
- Image classification
- Data preprocessing and feature extraction
- Model evaluation and performance optimization
- Team collaboration and problem-solving
Skills Gained:
Skills Gained:
- Network analysis
- Data visualization with Gephi
- Advanced analytical techniques
- Scenario-based failure analysis
- Collaborative problem-solving
Fine-Tuning BERT for Domain-Specific
Named Entity Recognition
Named Entity Recognition
2024
Our team fine-tuned pre-trained BERT model for Named Entity Recognition (NER) using data from J.R.R. Tolkien's Lord of the Rings universe. By adapting BERT with specialized datasets, we aimed to improve its accuracy in identifying domain-specific entities like characters, locations, and organizations. Training on both English Web Treebank and Lord of the Rings datasets led to significant performance improvements, demonstrating the value of incorporating domain-specific data.
Our team fine-tuned pre-trained BERT model for Named Entity Recognition (NER) using data from J.R.R. Tolkien's Lord of the Rings universe. By adapting BERT with specialized datasets, we aimed to improve its accuracy in identifying domain-specific entities like characters, locations, and organizations. Training on both English Web Treebank and Lord of the Rings datasets led to significant performance improvements, demonstrating the value of incorporating domain-specific data.
Second Year of Bachelors
2024
2024
Data Visualisation and Data-driven Decision Making
Natural Language Processing and Deep Learning
Visualization Project: WWII Casualties Analysis
2024
In this project, I collaborated with my classmates to develop a timeline visualization to represent Soviet casualties in WWII, using a mountain-shaped graph to highlight the peaks during key battles. This approach allowed viewers to quickly grasp the scale and timing of significant military losses. We chose red to symbolize the urgency and sacrifice associated with these events, while black conveyed the somber tone of the subject matter. The visualization not only aimed to inform but also to create an emotional connection with the data.
In this project, I collaborated with my classmates to develop a timeline visualization to represent Soviet casualties in WWII, using a mountain-shaped graph to highlight the peaks during key battles. This approach allowed viewers to quickly grasp the scale and timing of significant military losses. We chose red to symbolize the urgency and sacrifice associated with these events, while black conveyed the somber tone of the subject matter. The visualization not only aimed to inform but also to create an emotional connection with the data.
- Natural Language Processing (NLP)
- BERT fine-tuning and model training
- PyTorch framework
- Named Entity Recognition (NER)
- Data preprocessing and tokenization
- Performance evaluation (F1 score, precision, recall)
Skills Gained:
Skills Gained:
- Data Visualization
- Color Theory and Symbolism
- Narrative Visualization and Data Storytelling.
- Historical Data Interpretation
- Technical Tools (Excel, Tableau)
Third Year of Bachelors
2024/2025
2024/2025
In Progress
My Technical Projects
Alongside my Bachelor's projects, I have completed several technical projects aimed at improving public service efficiency and enhancing IT infrastructure. These projects, undertaken across the entire Tallinn library system, gave me valuable hands-on experience in system automation, hardware installation, and large-scale IT management.
Computer Upgrades: Drives, RAM & Network Cards
2024
In this project, I managed a large-scale hardware upgrade for over 270 standard desktop computers and 150 Thin Clients. I replaced hard drives, network adapters, and RAM to extend the machine’s lifecycle and improve performance. This required careful handling of both standard and compact hardware configurations. The upgraded computers were optimized for better connectivity, processing speed, and user performance, ultimately supporting a more reliable IT infrastructure.
In this project, I managed a large-scale hardware upgrade for over 270 standard desktop computers and 150 Thin Clients. I replaced hard drives, network adapters, and RAM to extend the machine’s lifecycle and improve performance. This required careful handling of both standard and compact hardware configurations. The upgraded computers were optimized for better connectivity, processing speed, and user performance, ultimately supporting a more reliable IT infrastructure.
Self-Service Library Kiosks, Book Scanners
& Automated Book Delivery Machine for Tallinn Libraries
& Automated Book Delivery Machine for Tallinn Libraries
2024
As part of a digital transformation initiative for the entire Tallinn library system, I implemented self-service kiosks, book scanners, and an automated book delivery machine. These installations streamlined library operations by allowing patrons to check out and return books independently, reducing wait times and easing the load on library staff. The project required configuring the kiosks to integrate seamlessly with the library’s existing systems, ensuring smooth data flow and secure book handling. This user-focused solution helped to modernize library services and improve overall accessibility for the community.
As part of a digital transformation initiative for the entire Tallinn library system, I implemented self-service kiosks, book scanners, and an automated book delivery machine. These installations streamlined library operations by allowing patrons to check out and return books independently, reducing wait times and easing the load on library staff. The project required configuring the kiosks to integrate seamlessly with the library’s existing systems, ensuring smooth data flow and secure book handling. This user-focused solution helped to modernize library services and improve overall accessibility for the community.
Technical Projects
- Hardware Replacement
- Thin Client Configuration
- Large-Scale Technical Project Management
Skills Gained:
Let’s Talk Data and Innovation
Skills Gained:
- Public Service Kiosk Installation
- Technical Troubleshooting
- Hardware Integration with Library Management Systems
- Project Management in Public Spaces
Learn More About Me