The study, using Jakob Nielsen’s Usability Principles, aimed to identify usability challenges in the Canvas LMS. Conducted by University of North Texas computer science students familiar with usability evaluation, it selected students for the sample. The research method served as a pilot, guiding future investigations. Ten students participated, with the questionnaire’s Cronbach’s alpha at 0.969, showing high reliability and consistency in the tools used.

Twitter Accounts Suggestion is concerned with recommending accounts for the users according to their tweet contents. Twitter contains a massive amount of data that can be useful for knowing each user’s preferences. This paper uses Named Entity Recognition (NER), one of the techniques used in Natural Language Processing (NLP). We propose a pipeline technique to analyze the textual content of tweets to recommend the proper user accounts (people or organizations) to facilitate mentioning other user accounts in the tweets.

The difference between minors and adults is an important legal distinction codified within multiple legal frameworks. Moreover, the distinction is important information to extract from unstructured text when applying natural language processing tasks such as anonymization, and abuse detection. Despite the value, identifying minors remains a manually performed task of which little research has been dedicated and no benchmarks are established. This paper seeks to address this need with two goals. First, the creation of labelled and publicly available dataset. Second, to demonstrate proof-of-concept with a BERT named entity recognition model, fine-tuned to make the subject distinction based on surrounding context. Toward this end, we created a custom dataset consisting of 2,534 sequences with 8,770 labelled instances of minors. This data was used to fine-tune a BERT named entity recognition model which demonstrated context aware tagging of minors versus adults. The model achieved an 89% f1 score in detecting minors and a 61% f1 score in detecting adults. While the performance of this model may not be suited for a production environment, we have established a starting point for future research in specific privacy protections for minors within a data processing stack.