Integrating LSA-based hierarchical conceptual space and machine learning methods for leveling the readability of domain-specific texts

Abstract

Text readability assessment is a challenging interdisciplinary endeavor with rich practical implications. It has long drawn the attention of researchers internationally, and the readability models since developed have been widely applied to various fields. Previous readability models have only made use of linguistic features employed for general text analysis and have not been sufficiently accurate when used to gauge domain-specific texts. In view of this, this study proposes a latent-semantic-analysis (LSA)-constructed hierarchical conceptual space that can be used to train a readability model to accurately assess domain-specific texts. Compared with a baseline reference using a traditional model, the new model improves by 13.88% to achieve 68.98% of accuracy when leveling social science texts, and by 24.61% to achieve 73.96% of accuracy when assessing natural science texts. We then combine the readability features developed for the current study with general linguistic features, and the accuracy of leveling social science texts improves by an even higher degree of 31.58% to achieve 86.68%, and that of natural science texts by 26.56% to achieve 75.91%. These results indicate that the readability features developed in this study can be used both to train a readability model for leveling domain-specific texts and also in combination with the more common linguistic features to enhance the efficacy of the model. Future research can expand the generalizability of the model by assessing texts from different fields and grade levels using the proposed method, thus enhancing the practical applications of this new method.

A study on the effects of language and visual art integrated teaching on language learning performance and satisfaction of ethnic minority students in China

“…Moreover, this approach considers ethnic minority students as the subjects of the curriculum and emphasizes individual natural development while improving the educational value of art activity. The instructors select topics for integrating into the curriculum from various perspectives in order to contribute to ethnic minority students' development in various dimensions during the participation in the course (Tseng et al, 2019). The research hypothesis, namely H1: Language and visual art integrated instruction has significant effects on language satisfaction, is supported, which conforms to the research results of Hasan et al (2021), Rahmatika et al (2021), andSantos andCastro (2021).…”

Construction and validation of a computerized creativity assessment tool with automated scoring based on deep-learning techniques.

“…Perhaps the most famous example is that the word “queen” can be derived from the vector-space word representations of “king,” “man,” and “woman”: The result of the vector subtraction of vec(“king”; i.e., the word vector of “king”) from vec(“man”) and the vector addition of vec(“woman”) is closer to vec(“queen”) than to any other word vector. More recently, researchers have used Word2Vec-based algorithms to identify linguistic relations beyond the word level (i.e., relations at the sentence, paragraph, and discourse levels), with initial successes being reported (Hsu et al, 2018; Tseng et al, 2019; Wang et al, 2016). Some studies have obtained effective results by grouping synonymous phrases and classifying short texts (e.g., Bansal & Srivastava, 2018; Zhang et al, 2015).…”

Supervised and Unsupervised Neural Approaches to Text Readability

“…Sequential Minimal Optimization (SMO) classifier with linear kernel achieved the classification accuracy of 78.13% for three readability classes (elementary, intermediate, and advanced reading level). An even more recent approach to readability classification conducted on Taiwanese textbooks was proposed by Tseng et al (2019). The main novelty of the research was the introduction of a latent-semantic-analysis (LSA)-constructed hierarchical conceptual space that can be used as a feature for training an SVM classifier for domainspecific readability classification.…”