Cloze-style queries are representative problems in reading comprehension. Over the past few months, we have seen much progress that utilizing neural network approach to solve Cloze-style questions. In this paper, we present a novel model called attention-over-attention reader for the Cloze-style reading comprehension task. Our model aims to place another attention mechanism over the document-le...

Computer “reading” of natural languages is a growing research topic with numerous applications, ranging from areas of artificial intelligence to linguistics. A successful “reading” of natural languages can be interpreted in two ways: 1) extracting facts out of a passage and 2) capturing “meaning” of a text. We focus on the latter interpretation, defining “meaning” as is done in humans, namely, ...

Cloze was officially introduced in a journal on Journalism as a technique for estimating text readability and as "a new psychological tool for measuring the effectiveness of communication" (Taylor, 1953: 415). Different varieties of cloze have since been developed and experimented upon as measures of such diverse traits as reading comprehension and language proficiency. The findings of numerous...

O teste de Cloze é amplamente utilizado na avaliação da compreensão leitura escolares. Estudos indicam que os resultados do por razão fixa se relacionam com a escrita, sugerindo esse formato dependente dessa habilidade linguística. Destarte, o objetivo desta pesquisa foi investigar interferência dos erros escrita para desempenho no fixa. Participaram deste estudo 1842 alunos, provenientes das c...

We describe a method to semi-automatically generate incorrect choices, or distractors, for cloze (fill-in-the-blank) questions. We generated distractors aimed at revealing what type of misunderstanding a student was having. English as a Second Language learners answered a series of cloze questions that presented distractors generated by our method. We analyzed their answers in order to see how ...

This paper describes an ensemble system submitted as part of the LSDSem Shared Task 2017 the Story Cloze Test. The main conclusion from our results is that an approach based on semantic similarity alone may not be enough for this task. We test various approaches and compare them with two ensemble systems. One is based on voting and the other on logistic regression based classifier. Our final sy...

We use vector space models (VSMs) as explicit models of word relations that influence the N400, and use this connection to predict N400 amplitude in an ERP study by Federmeier and Kutas (1999). We find that the VSM-based model is able to capture key elements of the authors’ manipulations and results, accounting for aspects of the results that are unexplained by cloze probability. This demonstra...

Participants read sentences presented one word at a time, half of which ended with a semantically incongruent ending. 1.5T functional magnetic resonance imaging data were collected from 11 participants, showing that the left posterior inferior temporal region, which has previously been termed the Language Formulation Area (LFA), responded to cloze probability. It is suggested, based on anatomic...

The notion of expectancy grammar as a key to understanding the nature of psychologically real processes that underlie language use is introduced by Oller (1979). A central issue in this notion is that expectancy generating systems are constructed and modified in the course of language acquisition. Thus, one of the characteristics of language proficiency is that it consists of such an expectancy...