Contextual Text Analysis for Assessing Reading Comprehension in Digital Education Tools
Abstract
The reading comprehension of digital education tools is the area of assessment that can be used to determine the cognitive aspect of learning text. The contextual text analysis will give information on how students receive and process information in the digital learning context. Conventional forms of assessment, including multiple-choice test or matching a key word, do not usually reflect in-depth learning, contextual associations, and the complexity of language interpretation and thus restricts individual commentaries. In a bid to overcome these shortcomings, a Bidirectional Encoder-Decoder Architecture (BEDA) is proposed in this paper. The BEDA model is a two-way process of student responses and captures the pre-contextual and post-contextual information and produces semantic representations to assess the understanding. With the addition of this structure to digital educational programs, teachers have the opportunity to check the level of understanding automatically, give feedback in real time, and change the level of content complexity depending on the specifics of a specific learner. The outcomes of experiments prove that the proposed method increases the effectiveness of assessing comprehension, detects learning gaps, and allows individual learning strategies.