Welcome!

This is the blog of Stefan Savev. I build software for data processing with machine learning, big data, backend and DevOps technologies. I hope you enjoy the content and learn something interesting and useful.

Machine Learing, Search and Big Data Algorithms

Methods for Efficient Structuring and Searching of Sparse and Dense Data (>= 9 posts on cosinesimilarity.org)
Random Trees for Nearest Neighbor Search (pdf)
Divergence from Randomness for Exploratory Data Analysis (on linkedin.com)
Three Learning Mechanisms in Neural Nets: a Different Perspective (on linkedin.com)
Fast Data Mining Of Related Phrases
Hierarchical Clustering That Works
Deriving Autosuggest From Raw Text Data
Searching Inside Covid 19 Papers using Keyword Classifiers
The Intuition Behind t-SNE

Efficient Algorithms

How to be 40 Times Faster than the Obvious Approach
Fast String Sort in C# and F# (on www.codeproject.com)
Fast External Sort (on https://www.codeproject.com)
References on the Random Projection method

Abstraction and Domain Specific Languages

Metalinguistic Abstraction (linkedin.com)
Your Cloud Infrastructure is a Graph (linkedin.com)
Functional Programming with F# for Data Analysis and Big Data Tooling (pdf, from 2009)

Stock Market Analysis

Open Source Repository Using The Deutsche Börse Public Dataset (on github.com). Featured on AWS opendata registry.
Picking Stocks Based on Volume (on linkedin.com and quantiopian.com)