Martin Frešer

Data Scientist at Sportradar
  • Claim this Profile
Contact Information
us****@****om
(386) 825-5501
Location
Slovenia, SI
Languages
  • English -
  • Slovene -
  • Croatian -

Topline Score

Topline score feature will be out soon.

Bio

Generated by
Topline AI

5.0

/5.0
/ Based on 1 ratings
  • (1)
  • (0)
  • (0)
  • (0)
  • (0)

Filter reviews by:

Nino Požar

Data science, machine learning, business intelligence or even volleyball, you name it and Martin has already mastered it. We’ve joined our hands on many projects, and Martin is one of the best people I have worked with. He makes sure all the deadlines meet and that also with the highest standards. His work ethics are pristine, and he is easily adjustable to a given situation. He has expertise and knowledge in the field, and the way he handles the clients is genuinely remarkable. Martin is very friendly and communicative, and many people in the workplace find his enthusiasm and dedication, both inspiring and motivating. He has earned everybody’s respect through his skills and value. He is smart and handles every situation very carefully and in the right manner. I consider myself lucky to work with such a professional. In any project you would be lucky to have Martin in. I most definitely recommend working with Martin!

You need to have a working account to view this content.
You need to have a working account to view this content.

Credentials

  • Microsoft: DAT275x Principles of Machine Learning: Python Edition
    edX
    Dec, 2018
    - Nov, 2024

Experience

    • Switzerland
    • Information Services
    • 700 & Above Employee
    • Data Scientist
      • Dec 2022 - Present

    • Austria
    • IT Services and IT Consulting
    • 500 - 600 Employee
    • Data Scientist
      • Oct 2016 - Feb 2023

      My job in data science team is to find the needle (meaningful projects) in a haystack of data of our customers and deliver solutions based on it, which boost and even change their decision-making. The goal is to make companies realize the benefits of becoming data-driven. Some of the projects I work(ed) on: o Implemented software for Production scheduling, powered by evolutionary algorithms for optimized plan, and carefully designed user interface o Machine learning platform for the leading Slovenian distributor. It optimizes purchase lifecycle according to forecasts of sales and automatically outputs daily purchase orders, which results in optimized stock level on portfolio of several thousand items o Customer segmentation with ML-powered algorithms for hospitality business I am working with fast changing and very powerful technologies: o Open source (Spark, Python libraries like sklearn, pandas (favourite), keras) o Self-written custom algorithms, which solve unique problems for different industries (sadly there is still no free lunch for all) o Cloud technologies (Azure with Databricks). o BI tool QlikSense enhanced with custom extensions (JavaScript) and backend (Django). Show less

    • Slovenia
    • Research
    • 400 - 500 Employee
    • Researcher
      • Feb 2016 - Oct 2016

      Department of the Intelligent Systems (E9). We worked on the European AAL project Fit4Work. I was performing data analysis and machine learning techniques for predictions with environmental countinous signals (temperature, co2, humidity,noise...). In scope of the project, we sent and got accepted an article at the UbiComp (leading conference for the ubiquitous computing) conference workshop Smarticipation and poster at the conference..

    • Student Job
      • 2013 - Feb 2016

      Formerly I worked at E9 as a student. Along two years I was working on various European projects: Commodity12, where I was analysing sound signals and upon them predicting high level activities, I was analysing ECG signal and search for anomalies (e.g. arrhythmia). I wrote two articles on Jožef Stefan International Post Graduate School Student conference and Information society conferrence also held by Jožef Stefan Institute.

    • Slovenia
    • IT Services and IT Consulting
    • Data Scientist
      • Feb 2015 - Aug 2015

      We worked on a stock trading project with a big set of historic data on stock trades. In short, we predicted, whether the stock trade was beneficial or not in a given time frame. I was responsible for evaluating various machine-learning algorithms, and to implement novel ensemble method Meta-DES. We evaluated models on back-test for 2 years and ensemble method has shown best results along all the tested algorithms. We programmed in R and Python languages. We worked on a stock trading project with a big set of historic data on stock trades. In short, we predicted, whether the stock trade was beneficial or not in a given time frame. I was responsible for evaluating various machine-learning algorithms, and to implement novel ensemble method Meta-DES. We evaluated models on back-test for 2 years and ensemble method has shown best results along all the tested algorithms. We programmed in R and Python languages.

Education

  • University of Ljubljana, Faculty of Mathematics and Physics
    Master’s Degree, Mathematics and Computer Science
    2013 - 2016
  • University of Maribor, Faculty of natural sciences and mathematics
    Bachelor’s Degree, Mathematics
    2010 - 2013

Community

You need to have a working account to view this content. Click here to join now