Skip to main content

Master's theses


In 2021, the students of the third intake of the Industrial Master's Programm in IT defended a total of seven master's theses:

  1. Brandon Christopher AutreyCustomer Journey Analysis at Pipedrive: A Process Mining Approach 
    Partner organisation: Pipedrive
    Supervisors: Marlon Dumas, Ambar Raj
  2. Gunel IsmayilovaAdopting Devops Practices: A Case Study 
    Partner organisation: Singularity Creations
    Supervisors: Ezequiel Scott, Madis Kapsi
  3. Karmen KinkClassification of E-Commerce Products Based on Textual Product Descriptions 
    Partner organisation: STACC
    Supervisors: Kairit Sirts, Karl-Oskar Masing
  4. Taavi LuikDesigning a Pharmacogenetic Test as a Medical Software Device 
    Partner organisation: UT Bioinformatics and Information Technology (BIIT) research group
    Supervisors: Sulev Reisberg, Kersti Jääger
  5. Ida Maria OrulaThe Process of Creating a Scientific Knowledge Base for Pharmacogenetic Testing 
    Partner organisation: UT Bioinformatics and Information Technology (BIIT) research group
    Supervisors: Sulev Reisberg, Kersti Jääger
  6. Aleksander PareloDevelopment of a virtual printer and print driver for Print in City 
    Partner organisation: Overall
    Supervisors: Urmas Tamm, Meelis Roos
  7. Tetiana ShtymTraffic light detection by fusing object detection and map info 
    Partner organisation: Bolt / UT Autonomous Driving Lab
    Supervisors: Tambet Matiisen, Meelis Kull


In 2020, the second intake of the Industrial Master's Programme in IT defended their Master's theses:

1. Kerstin Äkke and Datel completed a master's thesis entitled "Snow Cover Detection in Estonia from SAR Images Using Machine Learning Methods". The aim of this study was to test applicability of a method that combines most common features for snow detection extracted from SAR images in a machine learning model.
Supervisors: Viacheslav Komisarenko, Anti Gruno


Kerstin'si and Anti's comment on the Master's thesis: 

Kerstin's comment on the master's thesis:
"The topic of snow detection is of interest to the scientific community. The supervisor suggested this topic because working with SAR (Synthetic Aperture Radar) data is something that Datel is already working on as part of the Sille project. The topic itself is not directly related to any of these projects. Its main purpose was to support future work on SAR data. "

Datel Data Scientist Anti Gruno, supervisor:
“This Master's thesis has great practical value. The experience gained during the work, such as ESThub workflow modelling, object identification based on different parameters, machine learning will be used in various SAR development projects in the future.”[/collapsed]

[collapsed class="collapsible-blue"]

2. Alar Leemet joined Pipedrive through the Industrial Master’s Programme in IT and they found a mutual interest in the study of hackatons within Pipedrive. The aim was to find aspects affecting project continuation in a corporate setting using observations, surveys and interviews. The master's thesis is named "Hackathons for Corporate Innovation – a Longitudinal Study".
Supervisor: Alexander Nolte


Marko's comment on the Master's thesis:

Pipedrive’s Engineering Manager Marko Nõu said about the thesis:
“Why should any company organize hackathons? Are there any benefits? What about the projects created there? Alar's thesis covers this all and in an especially relevant aspect in todays situation - what about remote? Hackathons in Pipedrive have become almost axiomatic, but nobody has done such large scale research on them. There's lots of food for thought for participating teams as well for organizers.”[/collapsed]

[collapsed class="collapsible-blue"]

3. Kiryl Lashekvich was another Industrial Master at Pipedrive in the 2019-2020 period. He researched Pipedrive’s journey from Scrum-based practices to Mission-based in a master’s thesis called “Improving Agile Processes with Customized Mission-based Practices. Case Study”.
Supervisors: Marlon Dumas, Eerik Muuli


Marko's comment on the Master's thesis:

Here’s what Pipedrive’s Engineering Manager Marko Nõu said about the thesis:
“There has never been such comprehensive research into our software development process. Kiryl's thesis provides a great history lesson in retrospective from one side, but also deep insights and comparison of different processes we have used and their performance. It is a really useful summary and assurment that we are on the right track.”[/collapsed]

[collapsed class="collapsible-blue"]

4. Laura Ruusmann worked on her master's thesis "Comparison of Category-level, Item-level and General Sales Forecasting Models" at STACC. Laura tested out multiple methods for achieving best forecasting accuracy with lowest computational requirements: a traditional statistical forecasting approach (ARIMA), classical machine learning techniques (specifically ensemble methods) and a third one based on deep learning methods (specifically recurrent neural networks with LSTM architectures).
Juhendajad: Marlon Dumas, Eerik Muuli

5. Mohammad Mahdi Mohebbian partnered up with Positium and their mutual interest was to research "Real-time mobile data event detection system". In this thesis implementation of an enterprise system has been demonstrated for monitoring the behavior of the cell towers under the administration’s authority. The core functionality of this system is detecting ongoing events in different areas on an hourly-basis schedule utilizing multiple statistical approaches for abnormality detection.
Supervisors: Amnir Hadachi, Erki Saluveer


Mahdi’s comment on the master's thesis:

Mahdi’s comment on the master's thesis:
"Mastering knowledge and concept retrieval from big data was one of my primary motivations to join the Computer Science program in the first place. The idea of the "Event Detection System using mobile big data" was proposed by Erki Saluveer the CEO of Positium company during idea brainstorming not long after Metallica's concert in Tartu, The concert was essentially the trigger of the idea. I got to implement the enterprise using the most interesting big data manipulation tools and techniques. I enjoyed every single step of this journey and learned more than I thought I would." [/collapsed]

[collapsed class="collapsible-blue"]




In 2019 the very first intake of the Industrial Master’s Programme in IT defended their master's thesis:

  1. Maksym Semikin and STACC partnered up and researched a topic called "Jointly Tackling User and Item Cold-start with Sequential Contentbased Recommendations". This work presents a novel architecture for context-aware item prediction based on embeddings. The model combines item embeddings within a sequence to dynamically predict an item embedding for the next interaction. This allows to incorporate new items without model retraining.
    Supervisors: Tambet Matiisen, Carlos Bentes
  2. Simona Micevska joined Reach-U and together they found mutual interest in "A Statistical Drift Detection Method". In this thesis, the importance of interpretability in drift detection is highlighted and the Statistical Drift Detection Method (SDDM) is presented, which detects drifts in fast-evolving data streams with a smaller number of false positives and false negatives when compared to the state-of-the-art, and has the ability to interpret the cause of the concept drift. The effectiveness of the method is demonstrated by applying it on both synthetic and real-world datasets.
    Supervisors: Sherif Sakr, Toivo Vajakas
  3. Oleksandr Shvechykov worked on his master’s thesis in partner with Pipedrive and his research topic was "Scaling Up a Software Product: The Journey of Pipedrive". The thesis reflects on the Pipedrive's journey from a monolithic architecture to a microservices architecture, and from a static structure consisting of teams with fixed areas of responsibility to a dynamic structure consisting of dynamic teams with time-bounded missions.
    Supervisor: Marlon Dumas
  4. Margarit Shmavonyan worked on her master’s thesis "Improving Application Lifecycle Management at Swedbank: A Case Study" at Swedbank. The research work conducts a case study where ALM is examined from the Swedbank viewpoint which has expertise in managing applications. The study aims to reveal the issues and risks Swedbank is facing during the ALM process and find solutions for improving their ALM. As a result, a new approach was suggested and a detailed description of the flow was presented.
    Supervisors: Fredrik Payman Milani, Margus Melsas



Enterprises and institutions are invited to register for the Delta Career Day


The first AIRE Club brought together manufacturers and providers of support services



In 2021, we launched a dedicated newsletter to those interested in the industry collaboration programs, events and projects of the Institute of Computer Science.