In this chapter, the authors examined reading evaluation methods for foreign language learners based on learners’ reading processes. The goal of this chapter is twofold. The first is to evaluate text reading, and the other is to evaluate sentence reading. First, the authors assessed a text reading test to evaluate reading proficiency based on reading process, that is, effective reading speed, which is a complex measure of reading speed and comprehension rate. Statistical analysis confirms the adequacy of our effective reading speed test. Next, they propose a reading time model for evaluating reading proficiency at the sentence level. Their reading time model predicts sentence reading time based on the linguistic properties of a sentence and a learner’s proficiency. Linguistic properties consist of lexical, syntactic and discourse properties. Learners’ proficiency is defined using their score on the Test of English for International Communications (TOEIC). Their reading time model resulted in high prediction accuracy. From these results, they conclude that the reading process-based evaluation method is a promising test for foreign language reading proficiency.

No abstract

No abstract

This paper presents a method of automatically evaluating the fluency of machinetranslated sentences. We constructed a classifier that would distinguish machine translations from human translations, using Support Vector Machines as machine learning algorithms. In order to obtain a clue to the distinction, we focused on literal translations (word-for-word translations). The classifier was constructed based on features derived from word alignment distributions between source sentences and human/machine translations. Our method employed parallel corpora to construct the classifier but required neither manually labeled training examples nor multiple reference translations to evaluate new sentences. We confirmed that our method could assist evaluation on system level. We also found that this method could identify the †

No abstract