Welcome to
Anders Baumann's Website
I'm a data science graduate and data analyst focused on solving complex meaningful problems. I leverage data and AI to generate meaningful insights and drive impactful solutions.
MY PROJECTS
I'm a data science graduate and data analyst focused on solving complex meaningful problems. I leverage data and AI to generate meaningful insights and drive impactful solutions.
MY PROJECTS
I'm Anders Baumann, a data enthusiast specializing in data analysis and machine learning. I'm currently based in Oslo, Norway.ABOUT ME
I graduated from Menlo College, a business school in Atherton, California, USA, with a BSc in Business Analytics in May 2022, and I completed an MSc in Data Science for Business at BI Norwegian Business School in June 2024. My goal is to leverage machine learning processes to bring greater value to firms. I'm also an avid golfer who played collegiate golf for four years with scratch handicap.
Outside of work, I'm also an avid golfer, having played collegiate golf for four years with a scratch handicap.
Here are some of the things I like working on.
I am eager to learn both the theoretical theory of ML and its real life applications.
Conformal predictions provides reliable uncertainty estimates in predictive models, which is crucial for making informed, data-driven decisions in real-world applications.
I like telling stories with data, and attempting to solve problems through data-driven decision making.
I like to work in groups and collaborate with peers to learn from each other and produce greater work.
I like experimenting around with neural networks. It's incredible how these simple mathematical operations can create such beatiful outputs.
Data doesn't solve all business problems (yet) and it has its limitations. Therefore, it's important to know what it can answer, as well as what the data cannot answer.
Here are some of my works I'm most proud of.
This project is based on Spotify's REST API, and contains two programs. The first one tries to predict whether a song will be featured on the Billboard Hot 100 based on its audio features. The F1 score is 0.33, and was concluded to be difficult for even the more complex classification models. The precision score is 0.22, so it would not be feasible to use this in a real business setting. The other program recommend songs based on content-based filtering, again solely based on audio features given by the Spotify API.
My thesis consisted of evaluating the effectiveness of different Nautral Language processing (NLP) sentiment models in predicting the inflation expecations of economic agents based on the Federal Open Market Committee (FOMC) monetary policy statements. We provided two new fine-tuned LLMs trained specifically for deriving inflation sentiment of FOMC minutes and Bank of England Monetary Policy Reports. We compared with the SOTA LLM at the time base GPT-4o, and our fine-tuned BERT model outperformed this model in both evaluations. This model is freely available on Huggingface and can be accessed by clicking this text.
This program is made for golfers to understand their golf shot pattern better. The golfer sends in their golf shot data (distance, direction, and frequency), and my program estimates where the next golf shots will end up based on these previous shots. This program uses k-means-clustering to segment the different areas golf shots would likely end up. In addition, the algorithm to calculate the probability that a golfer will hit in a specific area uses among other techniques, the angle of the club face and the golfer’s competence level. The program was deployed as a REST-API linked to a website I developed. This website reached 200 unique visitors monthly.