Large amounts of information are nowadays easily obtainable using the Internet, and using implicit feedback whether a reader finds a text interesting is desirable. Eye-tracking technology could be used for such a feedback, and a combination of eye-movement features and a textual complexity measure can be used to predict the user's interest. In this paper we give an overview of a platform developed to evaluate and visualize implicit feedback of a person who reads a text. Based on the eye-movement samples provided, a model is trained that could be used to predict comprehensibility of a user reading a text. This prediction is combined with objective complexity evaluation of the text using data mining methods, and the outcome is used to select a text (from a repository) that a user may find more valuable (interesting). We briefly discuss the requirements, architecture and implementation of this platform.