Machine Learning for Holistic Evaluation of Scientific Essays

Hughes, Simon; Hastings, Peter; Britt, M. Anne; Wallace, Patricia; Blaum, Dylan

doi:10.1007/978-3-319-19773-9_17

Cited by 12 publications

(9 citation statements)

References 18 publications

(19 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this review, a few studies were identified, aiming to detect connections between key concepts or aim to group students' responses according to semantic similarities rather than lexical similarities. Although designed for science explanatory essay scoring, the structured learning approach using Recurrent Neural Networks (RNN) at the sentence and whole essay level is a promising future research area [29], [28], [59]. Other neural network models based on variations of RNN like LSTM [1], RNN with a co-attention module [65], or even convolution neural network [9] have been successfully implemented for automated essay scoring.…”

Section: Deep Learning Approachesmentioning

confidence: 99%

See 1 more Smart Citation

The State of the Art in Providing Automated Feedback to Open-Ended Student Work

Larrondo

Frank

Ortíz

2021

PCEEA

View full text Add to dashboard Cite

This article provides a review of the state of the art of technologies in providing automated feedback toopen-ended student work on complex problems. It includes a description of the nature of complex problems and elements of effective feedback in the context of engineering education. Existing technologies based on traditional machine learning methods and deep learning methods are compared in light of the cognitive skills, transfer skills and student performance expected in a complex problemsolving setting. Areas of interest for future research are identified.

show abstract

Section: Deep Learning Approachesmentioning

confidence: 99%

“…However, LSA and other similar approaches such as coherence index (Coh-Metrix, [48]) analysis are made at phrase or sentence level. Therefore, the authors [59] proposed a machine learning approach that combines the NLP system developed by Hughes [28] with LSA indices on similarity Coh-Metrix indices on coherence.…”

Section: Student Problem-solving Performancementioning

confidence: 99%

The State of the Art in Providing Automated Feedback to Open-Ended Student Work

Larrondo

Frank

Ortíz

2021

PCEEA

View full text Add to dashboard Cite

show abstract

“…The UNKNOWN token occurs frequently enough that it does not carry strong semantic content for the system, similar to words like "a" or "the". Then we applied our machine learning approach with a 7-word sliding window across the text to identify concepts within that window (Hughes et al, 2015). The fixed-size sliding window approach allows us to avoid the difficulties for machine learning from variable-length input, but the size of the window ensures that the words of a concept will almost always fall entirely within one of the windows (Hughes et al, 2015).…”

Section: Concept Detectionmentioning

confidence: 99%

“…Then we applied our machine learning approach with a 7-word sliding window across the text to identify concepts within that window (Hughes et al, 2015). The fixed-size sliding window approach allows us to avoid the difficulties for machine learning from variable-length input, but the size of the window ensures that the words of a concept will almost always fall entirely within one of the windows (Hughes et al, 2015). For each of the concept codes (the nodes in Figure 2), we trained a logistic regression classifier in which the features were the words and the bigrams within the window, as well as their relative positions within the window.…”

Section: Concept Detectionmentioning

confidence: 99%

Different Approaches to Assessing the Quality of Explanations Following a Multiple-Document Inquiry Activity in Science

Wiley

Hastings

Blaum

et al. 2017

Int J Artif Intell Educ

Self Cite

View full text Add to dashboard Cite

This article describes several approaches to assessing student understanding using written explanations that students generate as part of a multiple-document inquiry activity on a scientific topic (global warming). The current work attempts to capture the causal structure of student explanations as a way to detect the quality of the students' mental models and understanding of the topic by combining approaches from Cognitive Science and Artificial Intelligence, and applying them to Education. First, several attributes of the explanations are explored by hand-coding and leveraging existing technologies (LSA and Coh-Metrix). Then, we describe an approach for inferring the quality of the explanations using a novel, two-phase machine learning approach for detecting causal relations and the causal chains that are present within student essays. The results demonstrate the benefits of using a machine learning approach for detecting content, but also highlight the promise of hybrid methods that combine ML, LSA and Coh-Metrix approaches for detecting student understanding. Opportunities to use automated approaches as part of Intelligent Tutoring Systems that provide feedback toward improving student explanations and understanding are discussed.

show abstract

“…These subtasks related to comprehending evidence in relation to ideas, evaluating ideas in relation to evidence, and producing explanations based on evidence are complex and difficult. Most students have trouble with explanation and argumentation across all subject areas [6][7][8][9], but particularly with argumentation and explanation in science [10][11][12][13][14]. For instance, student explanations often focus on a single major cause to explain the scientific phenomena in question (i.e., lack completeness), while ignoring enabling and mediating factors (i.e., lack coherence) [15][16][17].…”

Section: Introductionmentioning

confidence: 99%

Engaging Middle School Students in Authentic Scientific Practices Can Enhance Their Understanding of Ecosystem Response to Hurricane Disturbance

McGee

Durik

Zimmerman

et al. 2018

Forests

View full text Add to dashboard Cite

Ecosystem response to hurricane disturbance is complex and multi-faceted. The likelihood of increased frequency of severe hurricanes creates a need for the general public to understand how ecosystems respond to hurricanes. Yet, opportunities to study disturbances to complex systems are rare in U.S. K–12 schools. Educators and researchers in the Luquillo Long-Term Ecological Research program used the results of research on ecosystem response to hurricane disturbance in the Luquillo Experimental Forest as a foundation for the development of Journey to El Yunque, a web-based, middle-school curriculum unit. The program engages students in using models as evidence to develop explanations for how particular species respond to hurricane disturbance. Prior research in education has shown that engaging students in a particular role, like that of a scientist, could have detrimental effects on students’ abilities to transfer what is learned from one context to another. In this research, we sought to understand whether having students engage in authentic scientific practices could support transfer of knowledge to the abstract context of a standards-based assessment. Students were randomly assigned to engage in the program in the role of a scientist or in the role of a student learning about an ecosystem. The dependent variables included students’ comprehension of the background readings, their predictions of population changes, and their overall learning of ecology. The results indicate that taking on a scientist role during the learning activities had an indirect effect on general ecology knowledge by increasing the quality of students’ notetaking during background reading. The results also indicate that students struggled to use their knowledge to develop a robust explanation for how species respond to hurricane disturbance. Journey to El Yunque provides a framework for engaging students in authentic investigations of hurricane disturbance. Future research will examine how to improve the quality of students’ final explanations.

show abstract

Machine Learning for Holistic Evaluation of Scientific Essays

Cited by 12 publications

References 18 publications

The State of the Art in Providing Automated Feedback to Open-Ended Student Work

The State of the Art in Providing Automated Feedback to Open-Ended Student Work

Different Approaches to Assessing the Quality of Explanations Following a Multiple-Document Inquiry Activity in Science

Engaging Middle School Students in Authentic Scientific Practices Can Enhance Their Understanding of Ecosystem Response to Hurricane Disturbance

Contact Info

Product

Resources

About