Automatic coding of students' writing via Contrastive Representation Learning in the Wasserstein space

Abstract

Qualitative analysis of verbal data is of central importance in the learning sciences. It is labor-intensive and time-consuming, however, which limits the amount of data researchers can include in studies. This work is a step towards building a statistical machine learning (ML) method for achieving an automated support for qualitative analyses of students' writing, here specifically in score laboratory reports in introductory biology for sophistication of argumentation and reasoning. We start with a set of lab reports from an undergraduate biology course, scored by a four-level scheme that considers the complexity of argument structure, the scope of evidence, and the care and nuance of conclusions. Using this set of labeled data, we show that a popular natural language modeling processing pipeline, namely vector representation of words, a.k.a word embeddings, followed by Long Short Term Memory (LSTM) model for capturing language generation as a state-space model, is able to quantitatively capture the scoring, with a high Quadratic Weighted Kappa (QWK) prediction score, when trained in via a novel contrastive learning set-up. We show that the ML algorithm approached the inter-rater reliability of human analysis. Ultimately, we conclude, that machine learning (ML) for natural language processing (NLP) holds promise for assisting learning sciences researchers in conducting qualitative studies at much larger scales than is currently possible.

VoxelEmbed: 3D Instance Segmentation and Tracking with Voxel Embedding based Deep Learning

“…The entire pixel embedding based learning consist of two major stages: embedding encoding and feature clustering. The RSHN network, combining convolutional GRUs [2,12] (ConvGRUs) and the stacked hourglass network [21], is employed as the backbone network for our voxel embedding. In [25], each pixel in a 2D video sequence was encoded to a high-dimensional embedding vector with the intuition that all pixels from the same cell, across spatial and temporal, should have the same feature representation (embedding).…”

GPU-accelerated Faster Mean Shift with euclidean distance metrics

“…Among the unsupervised clustering algorithms, mean-shift [13] is arguably the one of the most widely used clustering algorithm in clustering problems, which has been used in image segmentation [8], voice processing [14], [15], object tracking [9] and vector embedding machine learning [16]. The advantage of mean-shift is a density-based(centroid-based) clustering approach and can determine the number of clusters adaptively.…”

WIP: Using Machine Learning to Automate Coding of Student Explanations to Challenging Mechanics Concept Questions

“…Many studies also applied ensemble techniques like bagging, boosting on various text classification machine learning models to study student responses (Bertolini et al, 2021, Zhai et al, 2020a. Several studies have also used neural network models (Jiang et al, 2020;Luan et al, 2021;Rosenberg, 2021). However, to our knowledge, only a few studies for educational applications in general have leveraged Transformer-based machine learning models (Vaswani et al, 2017, Devlin et al, 2018, Raffel et al, 2019, Brown et al, 2020.…”