Essay Assessment with Latent Semantic Analysis

Abstract

Latent semantic analysis (LSA) is an automated, statistical technique for comparing the semantic similarity of words or documents. In this article, I examine the application of LSA to automated essay scoring. I compare LSA methods to earlier statistical methods for assessing essay quality, and critically review contemporary essay-scoring systems built on LSA, including the Intelligent Essay Assessor, Summary Street, State the Essence, Apex, and Select-a-Kibitzer. Finally, I discuss current avenues of research, including LSA's application to computer-measured readability assessment and to automatic summarization of student essays.

Extracting semantic representations from word co-occurrence statistics: A computational study

“…This has included using LSA to model metaphor comprehension (Kintsch, 2000;Kintsch & Bowles, 2002); a model of children's semantic memory built from an LSA analysis of a child corpus (Denhière & Lemaire, 2004); application to grading student essays (Miller, 2003); application of different sources of knowledge on reasoning (Wolfe & Goldman, 2003); mathematical improvements to the LSA distance measure (Hu et al, 2003); potential improvements in the statistical methods underlying LSA (Hofmann, 2001); and many other studies.…”

Automatic Scoring of an Analytical Response-To-Text Assessment

“…Baseline2 is LSA [9] trained on prescored essays and the text. While our first baseline came from the holistic scoring literature, LSA has been successfully used in trait-based systems to score content and ideas [8,12], which seems more similar to our task of scoring Evidence. Since we do not have a separate pre-scored set of training essays, we do cross-validation in our experiments.…”

“…Since we do not have a separate pre-scored set of training essays, we do cross-validation in our experiments. Scores are assigned based on the scores of the 10 most similar essays, weighted by their semantic similarity based on [12].…”

ReaderBench Learns Dutch: Building a Comprehensive Automated Essay Scoring System for Dutch Language

“…In addition, it should analyze also the quality of the text, that means its coherence and complexity. Latent Semantic Analysis (LSA) [1,2] was one of the first methods to introduce the possibility of measuring the semantic similarity when comparing a text written by a student to the corresponding learning base. Later on, Latent Dirichlet Allocation (LDA) [3] was introduced as a topic modeling technique that overcomes some problems of LSA.…”