Evaluation of Reading Support Tools by Reading Comprehension Tests and Reading Speed Tests

This paper introduces our reading process monitoring systems and also presents the experimental results that show the adequacy of our reading data. Our system divides a text into reading areas and records reading time for each area. We conducted two experiments using this tool to verify the adequacy of our reading process data. In the first experiment, we examined whether the reading process can distinguish easy text reading and difficult text reading, and confirmed the adequacy of our reading process data. In the second experiment, we tried to evaluate efficiency of reading support tools such as (i) chunker, (ii) glosser, and (iii) machine translation system, assuming that efficiency of these systems co-relates with text readability. The experimental results show that only the machine translation system effectively supports reading.