Automated Chinese Essay Scoring from Topic Perspective Using Regularized Latent Semantic Indexing

Abstract

Finding out an effective way to score Chinese written essays automatically remains challenging for researchers. Several methods have been proposed and developed but limited in the character and word usage levels. As one of the scoring standards, however, content or topic perspective is also an important and necessary indicator to assess an essay. Therefore, in this paper, we propose a novel perspective -topic, and a new method integrating topic modeling strategy called Regularized Latent Semantic Indexing to recognize the latent topics and Support Vector Machines to train the scoring model. Experimental results show that automated Chinese essay scoring from topic perspective is effective which can improve the rating agreement to 89%.

Predicting Evaluations of Essay by Computational Graph-Based Features

“…With the development of Chinese natural language processing technology (such as the maturity of Chinese word segmentation technology), a small number of Chinese AES research appeared in the early twenty-first century (Zhang and Ren, 2004;Li, 2006;Cao and Yang, 2007). Professor Liu and his team at Harbin Institute of Technology developed an automatic evaluation system for Chinese writing (Hao et al, 2014;Liu, 2015;Gong, 2016;Fu et al, 2018) and successfully implemented the scoring system on the national college entrance examination (NCEE), the Chinese proficiency test for minorities in China (MHK), and classroom writings. The features extracted by Chinese-oriented AES systems at first were mostly based on surface language features, such as word frequency (Zhang and Ren, 2004;Li, 2006).…”

Contextualized Latent Semantic Indexing: A New Approach to Automated Chinese Essay Scoring

“…Inspired by the studies for English, similar methods have been researched from various angles [16,20,26]. For instance, topic feature extraction methods using topic modeling strategies like regularized LSI (RLSI) [14,34] and latent Dirichlet allocation (LDA) [2,14] are novel ones. Other methods have also been tested; however, their results are relatively rare [4].…”

“…The first study of scoring Chinese essays from the topic perspective has been used in Ref. [14], with the batch version of RLSI. The target of RLSI is to decompose a term-document matrix D into a term-topic matrix U and a topic-document matrix V, minimizing the Frobenius norm 2 || || F − D UV with ℓ 1 -norm regularization on U and ℓ 2 -norm regularization on V, which can be described as…”

SCESS: a WFSA-based automated simplified chinese essay scoring system with incremental latent semantic analysis

“…The word level method concentrates on word usage purely, based on a word list trained by human-scored essays (Ke, Peng, Zhao, Chen and Wang 2011). Regularized Latent Semantic Indexing (RLSI) from the field of topic modeling focuses on topic(s) in a dataset (Hao, Xu, Peng, Su and Ke 2014;Wang, Xu, Li and Craswell 2013). Among them, LSA, designed for indexing documents for information retrieval, is the most common technique that has been successfully applied to a wide range of fields (Tonta and Darvish 2010;McInerney, Rogers and Jennings 2012;Wang and Yu 2009;Jin, Gao, Shi, Shang, Wang and Yang 2011), such as bioinformatics (Ismail, Othman and Kasim 2011), language processing (Wang and 2007; Yeh, Ke and Yang 2002;Gorrell 2006;Chang, Sung and Lee 2013) and signal processing (Mesaros, Heittola and Klapuri 2011).…”

“…2013; Hao et al . 2014). Among them, LSA, designed for indexing documents for information retrieval, is the most common technique that has been successfully applied to a wide range of fields (Wang and Yu 2009; Tonta and Darvish 2010; Jin et al .…”