The last module of the Study program of the Master's Degree University Degree in Big Data Science consists of the elaboration of a work that provides solutions to real problems and projects proposed by companies or institutions with which there are partnership agreements. In this way, the student acquires a very valuable experience internship in the field of data science and big data.

Lucía Colín Cosano is working on her TFM at the Swedish multinational company Ikea. "In companies like this, the amount of data that are collected daily is very voluminous, offering a great possibility to base the decisions that are taken."

The intake process of data is complex, so it is of vital importance to check the validity of these and to enrich those fields that are most relevant for generating useful models for the business.

"By ensuring the quality of the data we are able to make more accurate predictions, for which we use time series. In this way we can reduce the production start-up time of the processes, as well as know in advance the possible behavior of the market."

Jorge Fernández García is part of a team of 3 students who are developing their TFM with Sports Management managers and data analysis in a LaLiga soccer club.

"The data are related to everything around us and the world of soccer is no exception. The data analysis is the perfect partner for scouting".

The ability to collect and analyze data from players allows teams to obtain accurate information about their performance and skills, so they can make more informed decisions about their team composition and game strategies.

"At our work Final on Master's Degree we are finding that data analytics tools have become a factor core topic in the success of player scouting in today's soccer. In this way, club management is able to identify players with the greatest potential who might otherwise go unnoticed."

On the other hand, Elena Martín de Diego is working on her TFM in the high school of Science of the data and Artificial Intelligence (DATAI) at the University of Navarra, where she is a researcher. 

Decisions made by a machine learning algorithm have a great impact on people's lives. Moreover, it has recently been shown that such mechanisms can inherit existing patterns from data such as cultural inequalities between groups that differ in age, gender or ethnicity. 

This project has a twofold purpose. On the one hand, the review of repair techniques of the data in order to achieve unbiased algorithms; and, on the other hand, to highlight the power and flexibility of Python to address fairness issues. "The current research focuses its attention on developing a methodology in scenarios of data streaming"Elena assures.

Finally, Francisco Javier Alías, Data Scientist at department of the Global Risk Management Analytics Holding at BBVAbased his work on the study and implementation of syntheticdata in risk models.

The generation of synthetic data , traditionally associated with the world of imaging for applications such as augmented reality or medical imaging, has been able to achieve good results in problems where the size of the training set of data is smaller than that required by complex learning models, through the generation of new observations.

"The project is going to focus on exploring these methods and using them in risk models of credit . The techniques that will be studied and used range from simple algorithms to more complex Deep Learning models that try to learn the distribution of the set of available data ", says Francisco Javier.