Within the field of computer science (CS), women are under-represented in the workforce and education settings. As Massive Open Online Courses (MOOCs) grow in popularity, understanding the gender differences in reasons for enrolment and engagement remains crucial to improving learner outcomes. This study investigates why men and women enrol in introductory CS MOOCs and how they interact with these courses. This is done with data from four MOOCs offered by TU Delft between 2015 and 2022.

Using survey data for the learners' reasons for enrolment and clickstream data for their behavioural engagement, we applied k-means clustering to identify engagement patterns. Our analysis reveals that the three most important reasons for men and women are career-related, interest-related, and degree-related, in that order. Women are more likely to enrol for career-related reasons than men, while men are more driven by interest in the topic than women. Women also tend to show lower engagement levels compared to men, who are more likely to complete the courses. We found no significant association between reasons for enrolment and engagement for men and women.

These findings highlight the need for gender-sensitive course design strategies to enhance engagement and completion rates. Providing mentorship opportunities, fostering peer interaction platforms, and highlighting role models in the field could also help create a more inclusive learning environment. Future research should explore specific learner challenges and incorporate a more comprehensive engagement model.

This paper presents the findings of a benchmark performed on the audio fingerprinting framework OLAF in the context of movie music. The goal is to find a music identification framework suitable for automatically identifying a song from a movie clip. This research aims to find how well OLAF performs in this context with regard to the criteria determined in the benchmark and to see if the performance can be increased. The OLAF framework makes use of parameters, which are individually tweaked and benchmarked with synthesised data. These findings are then combined to find a set of parameters that is used to perform this same benchmark on real clips from movies. The found parameter setup slightly increased the performance, going from 1 true positive and 133 false negatives in the original setup to 5 true positives, 3 false positives and 126 false negatives.