Sverrir H. Davíðsson

Hagnýtar gervigreindarlausnir

A class that Corey and I made in the fall of 2024 where participants learn to use AI solutions effectively in their field, with a focus on ChatGPT

Further information:

• Course 1 Details
• Course 2 Details
• Course 3 Details
• Course 4 Details
• Course 5 Details

• Radio Interview at Rás 2
• Radio Interview at Rás 1

• Interview with Corey and Me

Close Project

THE MUSIC INDUSTRY

Musicians, lyrics, genres. How is it all connected?

Welcome to our project for the course 02805 Social graphs and interactions.

Our objective with this project is to analyze the music industry using data from Spotify and Genius. We want to try to visualize a network of musicians from all types of music, find hidden structure in the music industry, and try to gain new insights into the different genres as well as develop a tool to discover and explore new music. We want to see if we can relate the network structure to genres, lyrical sentiment, and other characteristics in the data. We will look into lyrical themes to identify patterns for certain topics and genres, and see how it has evolved over the years.

We gathered the data for the analysis using the web APIs from Spotify and Genius. You’re welcome to download the datasets using the button below and check out our Explainer Notebook, which shows the development process step-by-step. In addition, there is a brief description of the datasets below.

Close Project

Stereo image pointclouds

My attempt at creating a tool that uses two smartphones to generate and combine multiple 3D point clouds using OpenCV

Close Project

Texture synthesis

I implement a non-parametric texture synthesis method by A. Efros and T. Leung. as well as a modification. The difference between this version and the original is that the original needed to be provided with an image with the texture, one that has no missing pixels, while this modification looks instead at the non-missing pixels from the image itself to fill in the missing pixels. This change also means that we can take a noisy image where no non-noisy version exists and fill in the noisy pixels. This image does not need to be of a texture, as is demonstrated on the Github page.

Close Project

3D printing a mountain

I wanted to refresh my skills in Blender and decided to combine a few challenges in one small project. I wanted to turn a 3D elevation map into a 3D printed miniature. I picked a well known mountain I've hiked multiple times close to Reykjavík as the subject.

Close Project

Patent visualization with t-SNE

This project investigates the potential of using public patent data for early detection of promising companies for venture capital financing. It uses a bag-of-words representation of patent abstracts and t-SNE visualization to identify patent clusters. The goal is to create an interactive tool demonstrating the value of this approach in a venture capital context by examining prominent patent holders and distribution in specific technology subfields.

Close Project

Unsupervised leakage detection

Leakage detection system using machine learning and sensor data.

This project aims to develop an efficient leakage detection system by combining sensor data and machine learning algorithms. The system can help identify leakages, allowing for prompt action to prevent further damage and conserve water resources. The project involves data collection, feature engineering, model training, and two proof-of-concept case studies where time series data with several known leakages is used to validate the results.

Close Project