Empowering Interviewer Training : Web application version development report
Arokari, Olli (2023)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi-fe202402147318
https://urn.fi/URN:NBN:fi-fe202402147318
Tiivistelmä
This Master's thesis is about the development project of Empowering Interview Training (EIT), specifically its web application version, based on a previous EIT version written in VB.NET. The purpose of EIT is to conduct research about solving quality issues when it comes to interviewing suspected victims of child sexual abuse (CSA), which are part of criminal investigations. EIT is an interview training tool, and research is focusing on whether such a tool can reliably be used for training professionals in order to increase the quality of the interviews. Research has also explored if such a tool can be used to not only train, but also maintain a sufficient level of quality of the interviews.
The introductory part of the thesis presents the underlying issue regarding the quality of interviews with children suspected of being the victims of sexual abuse, and how research is being conducted on the use of EIT as means to improve the interview quality. Additionally, the introduction explains the issues regarding the previous version of EIT, and how a web application version could potentially solve many of them. The background chapter continues by presenting the previous version of EIT in more detail, and introduces how the training is conducted. The background chapter also introduces the reader to the basics of computer networking and web development.
The development of the web application is presented in three phases. The first phase presents the requirements, architecture, initial release, and challenges faced during the phase. The second phase focuses on further development of EIT through improvements, new features, and correcting issues discovered through use of the application. The second phase also explains the deployment of the application to Microsoft Azure, and how Japanese collaborators aided the development through feedback as a result of testing the application. The third phase presents an attempt at automating the recognition of question types using artificial intelligence, which ultimately became a proof of concept with everything in place except a properly trained machine learning model.
The thesis concludes with an observed status of the application in 2023, and recommendations for further development.
The introductory part of the thesis presents the underlying issue regarding the quality of interviews with children suspected of being the victims of sexual abuse, and how research is being conducted on the use of EIT as means to improve the interview quality. Additionally, the introduction explains the issues regarding the previous version of EIT, and how a web application version could potentially solve many of them. The background chapter continues by presenting the previous version of EIT in more detail, and introduces how the training is conducted. The background chapter also introduces the reader to the basics of computer networking and web development.
The development of the web application is presented in three phases. The first phase presents the requirements, architecture, initial release, and challenges faced during the phase. The second phase focuses on further development of EIT through improvements, new features, and correcting issues discovered through use of the application. The second phase also explains the deployment of the application to Microsoft Azure, and how Japanese collaborators aided the development through feedback as a result of testing the application. The third phase presents an attempt at automating the recognition of question types using artificial intelligence, which ultimately became a proof of concept with everything in place except a properly trained machine learning model.
The thesis concludes with an observed status of the application in 2023, and recommendations for further development.