// Hello there!
Born on March 8th, 1998, I come from Sestao, a medium-sized town in the Basque Country (Spain).
Before moving my career path to Finland, I completed my Bachelor’s degree in Economics at the University of the Basque Country (Spain). Subsequently, attracted by a more quantitative field of studies, I completed my Master’s in Statistical Data Analytics at Tampere University (Finland) in 2023.
Today, I am a Doctoral Researcher at the University of Oulu (Finland), sweetly immersed in the research field of empirical software engineering.
// My academic profile
Currently. I am a Doctoral Researcher in the Empirical Software Engineering in Software, Systems and Services (M3S) Research Unit within the Faculty of Information Technology and Electrical Engineering (ITEE) at the University of Oulu. I combine teaching assistance, research work within the scope of my doctoral studies as well as collaboration with academic researchers within the University of Oulu and other universities in Europe.
Expertise. My profile is close to data science and statistical data analysis. I strongly specialize in time series analysis, machine learning, and deep learning algorithm applications in multiple fields of science. My expertise plays an important role in answering the defined hypotheses in my research topic of Time Dependence in Empirical Software Engineering.
Experience. Given a short professional experience closely related to academic research after my Master’s, my previous works include statistical modelling and inference projects. My recent academic outputs have covered research articles related to the implementation of Cohort Studies in Empirical Software Engineering and the application of Statistical Time Series Analysis techniques within the topic of Code Technical Debt.
// About my research
Researchers often delve into the connections between different factors derived from the historical data of software projects. For example, scholars have endeavoured to explore associations among these factors. However, in recent years, researchers have reported that the research community still fails to consider the temporal interdependence among these variables. This issue arises from the difficulty of dealing with time dependency and using statistical methods that often cannot capture the temporal factor from the variables. Still, these types of dangerous considerations have been reproduced during previous years due to the lack of robust analysis methods. My goal through my research is to emphasize the consequences of ignoring time dependence in data analysis within current research and enhance the research on new methodologies to address this issue. During my doctoral studies, I plan to work on potential methodologies to demonstrate the impact of time when drawing conclusions from empirical software engineering data analysis methodologies.
// Attended conferences
// Published papers
- Comparing Multivariate Time Series Analysis and Machine Learning Performance for Technical Debt Prediction: The SQALE Index Case (2024) | Accepted at TechDebt’24 Poster papers
- Cohort Studies for Mining Software Repositories (2024) | Accepted at MSR’24 Tutorials
- Analyzing the Ripple Effects of Refactoring. A Registered Report (2024) | Accepted at ICSME’24 Registered Reports
- In Search of Metrics to Guide Developer-Based Refactoring Recommendations (2024) | Accepted at ESEM’24 Registered Reports
// Collaborations
- Does Microservices Adoption Impact the Development Velocity? A Cohort Study. A Registered Report (2023) | Accepted at ICSME’24 Registered Reports
// Planned research stages
- Applied Bayesian Statistics: It is one of the statistical theoretical bases of modern Machine Learning algorithms. Through the logic defined in the Bayes Theorem, the available knowledge about parameters in a statistical model is updated with the information in observed data to determine the posterior distribution. My aim is to implement core Bayesian algorithms and inspect their efficiency through the analysis of software engineering metrics anchored to the assumption of time dependence.
- Temporal Logic: These models differ by the ontological assumptions made about the nature of time in the associated models, by the logical languages involving various operators for composing temporalized expressions, and by the formal logical semantics adopted for capturing the precise intended meaning of these temporal operators. I plan to study this branch of the temporal analysis field through a research visit to Università Degli Studi di Milano – Bicocca (Italy) under the supervision of my PhD co-supervisor.
- Applied Transformers: These models have shown great ability to model long-range dependencies and interactions in sequential data and time series modelling. Many variants of Transformer have been proposed to address special challenges in time series modelling and have been successfully applied to various time series tasks, such as forecasting and anomaly detection.
// My message to the reader (and to me every day)
“Knowledge is an unending adventure at the edge of the uncertainty.”
Leto Atreides II (From the book “Children of Dune”)
