| The increasing popularity of Social Reference (SR) services has enabled a corresponding growth in the number of users engaging in them as well as in the number of questions submitted to the services. However, the efficiency and quality of the services are being challenged because a large quantity of the questions have not been answered or satisfied for quite a long time. In this dissertation project, I propose using expert finding techniques to construct an automated Question Triage (QT) approach to resolve this problem. QT has been established in Digital Reference (DR) for some time, but it is not available in SR. This means designing an automated QT mechanism for SR is very innovative.;In this project, I first examined important factors affecting triage decisions in DR, and extended this to the SR setting by investigating important factors affecting the decision making of QT in the SR setting. The study was conducted using question-answer pairs collected from Ask Metafilter, a popular SR site. For the evaluation, logistic regression analyses were conducted to examine which factors would significantly affect the performance of predicting relevant answerers to questions.;The study results showed that the user’s answering activity is the most important factor affecting the triage decision of SR, followed by the user’s general performance in providing good answers and the degree of their interest in the question topic. The proposed algorithm, implementing these factors for identifying appropriate answerers to the given question, increased the performance of automated QT above the baseline for estimating relevant answerers to questions.;The results of the current study have important implications for research and practice in automated QT for SR. Furthermore, the results will offer insights into designing user-participatory DR systems. |
