In social media research the lack of ground truth for evaluation is a recurrent problem. We study the preference mining task in Twitter network which suffers from this lack of ground truth problem. We implement three different methods from literature, considering a common preference domain of news and carry a comparative study among them. Our preliminary findings show that is possible to combine methods in order to avoid unfeasible user surveying baselines and enable the evaluation of techniques. In the future, our target is to completely eliminate ground truth sets and evaluate based on correlation and causality techniques.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.