Blogs And Projects

Finding correlation between heart rate and frustration

In this blog we examined the interaction between frustration levels and heart rate during two-person puzzle-solving tasks. We employed statistical tests to determine assumptions and thereafter conducted further testing to find that there was no statistically significant effect of frustration on the proxy heart rate metrics...

Read More

Regression and classification on french wine

In this blog, we delve into an insightful exploration of machine learning and data mining techniques applied to a wine dataset, aimed at predicting alcohol content and classifying wine cultivators. We harness a dataset comprising 178 wine instances from three different Italian cultivators, analyzed through features such as alcohol percentage, malic acid, and phenols. The study employs methods including linear regression with regularization and artificial neural networks (ANNs), underpinned by cross-validation for error estimation. Our findings...

Read More

AI Pacman with deep-Q-learning

This blog is about the game Pac-man's mechanics, the essential mechanic of avoiding a ghost in a maze. It is relevant, because if we can get an Ai to solve one mechanic of a game then we can build an Ai to solve multiple mechanics and problems. In this blog we have chosen to research, how well a deep-Q-network Ai can survive a single ghost dependent on different amounts of learning. Both the ghost and Ai agent will spawn in random locations and the ghosts movement is random as opposed to the real Pac-man game....

Read More

Natural Language Processing using FastText

This blog explores an academic endeavor involving the FastText method applied to a news reports dataset. The project's objective was to optimize and evaluate FastText for categorizing news into sports, world, business, and science & technology....

Read More

Data Processing of ETF's in R

his blog post explores a comprehensive statistical analysis project conducted on a dataset comprising weekly returns of 95 ETFs over a span of 454 weeks from May 2006 to May 2015. The analysis delves into data quality, empirical distribution, and performance metrics of selected ETFs, leveraging statistical methods...

Read More

Hand detector, used for egonomic office pratice

In the modern workspace, where digital interfaces dominate our daily tasks, the importance of ergonomic practices cannot be overstated. As we delve into the future of workplace health, a groundbreaking project emerges, promising to redefine our interaction with technology and mitigate the growing concern of musculoskeletal disorders among office workers...

Read More