Topics for Bachelor’s and Master’s Theses

Below is a list of topics for bachelor’s and master’s theses.
If you are interested in any of them, contact sandi.baressi.segota@unipu.hr.

Almost all topics can be used for both bachelor’s and master’s theses, depending on the complexity of the exact task we agree on.

Analysis and Classification of Network Traffic

Using Wireshark or a similar software package, monitor the network and store packet data (number of packets in a flow, total amount of data, average RTT, packet size, uplink/downlink ratio, number of protocols (TCP/UDP), variance of packet sizes, average inter-arrival time). Based on the collected data, use machine learning methods to develop a model that detects different types of traffic (e.g., interactive traffic or video traffic).

Development of a Data Transfer Protocol

Develop a protocol adapted for the transmission of a specific type of data, using HTTPS communication or similar. Adjust the protocol to the needs of that specific category of data. Develop and test program code, pseudocode, and a state machine representation.

Development of an Interactive Graph Simulator

Develop software that, based on an input table of graph connections, automatically tests multiple algorithms for finding the shortest path between two arbitrary points. Perform performance testing of the algorithms on networks of different sizes and experimentally determine which ones have the best performance.

Simulation of a Robotic Process

Choose a manufacturing, service, or any other process. Using RoboDK or CoppeliaSim, simulate the operation of a robot in that process and propose ways of integrating the robot into the system.

Modeling a Process/Device Using Machine Learning

Based on available information and equipment, create a dataset and conduct classical statistical analysis and modeling using suitable machine learning methods.

The World Dataset

For as many countries as possible, collect as many relevant values as possible (e.g., population, GDP…). Perform a basic analysis of the dataset.

Testing Data Synthesis Reversion

Based on real data, generate a synthetic dataset. Attempt to develop a GAN-based model that tries to regenerate the original data from the synthetic data.

User Movement Prediction

Detect the user’s hands in a video. Based on the motion, predict the user’s future movement and visualize it.

Testing Obstacle Avoidance

Develop a system that simulates obstacle detection on a robot and avoids it in several different ways. Compare which of the defined avoidance algorithms is most suitable in different environments.

Visualization of Potential Fields

Using a computer vision detection algorithm, determine the elements present in the scene. Using the potential fields method, determine the best path through the area. Test algorithm performance with respect to different parameter settings.