Portfolio

These are data science projects using technologies such as R, R Shiny, ggplot, plotly, linear regression modelling, and more.

Personal projects

These are data science projects I created for fun and/or in order to learn or expand my knowledge on the technologies used.

Shiny: TV viewing dashboard

This Shiny app is a dashboard for my viewed TV show episodes. It shows totals, averages, plotly plots, as well as raw data, which can all be filtered according to a variety of options.

Shiny: Book value analysis

In this Shiny project I analyze the books I read since Oct 2023 and how much money I spent on them, using plotly plots.

ggplot: Roleplaying Community Survey

As an organizer of the bi-monthly English-speaking Roleplaying Jam in Düsseldorf, I was curious about the demographics of this community and their language preferences when it comes to roleplaying. I ran a survey using SoSciSurvey and analyzed the data using R, creating ggplots.

ggplot: Analyzing Doctor Who (2005) Viewership

As a long time viewer of Doctor Who (2005), I was curious about the decline in quality of the show. In order to visualize this decline, I created ggplots in R of the IMDb ratings and UK viewership numbers of the show's Seasons 1 through 13, taking into account factors such as showrunner, Doctor and companions.

plotly.js: Robin Hobb Book and Reading Speed Analysis

After I finished reading a 16-book series by author Robin Hobb, I was interested in my own personal reading speed, so I analyzed the lengths of these books and came up with my own measure to analyze my reading speed based on book length, plotted in plotly.js.

Academic Research

These are data science projects I worked on during my time in academia. Most are part of my PhD project.

Paradigm uniformity in inflectional stems

My dissertation contains 5 different projects related to paradigm uniformity effects (see other projects on this site for details), and was published as a 168+ page book by düsseldorf university press.

Comprehension experiment using mousetracking

I conducted a mousetracking experiment to investigate whether listeners can perceive acoustic differences in words spliced from segments of word pairs like days and daze. Details of this study are available in Chapter 6.2 of my dissertation.

Same different experiment

This experiment investigated whether participants can hear an acoustic differences between stems of words (e.g. day in days) and artificially lengthened versions of these stems. Details of this study are available in Chapter 6.1 of my dissertation.

Production experiment

This experiment elicited target items by means of a cloze task. The target items were late analyzed under the assumption that there would be durational differences between monomorphemic and complex words. Details of this study are available in Chapter 5 of my dissertation.

Paradigm uniformity: QuakeBox Corpus Study

My published paper Phonetic reduction and paradigm uniformity effects in spontaneous speech investigates durational differences in homophonous monomorphemic and complex word pairs, using data from the QuakeBox corpus. I later replicated this study with the Buckeye Corpus (see Chapter 4.3 of my dissertation).

'Mate' in Australian English

Inspired by my semester abroad in Australia, I examined the gender restrictions of the address term 'mate' in Australian English. I conducted a survey study in 2016, which I replicated and expanded in 2022. It is scheduled to be published as In Australia, can anyone be your ‘mate’? - Gender restrictions of the address term ‘mate’ in Australian English.