Recent years have seen a flourishing of Natural Language Processing models that can mimic many aspects of human language fluency. These models harness a simple, decades-old idea: It is possible to learn a lot about word meanings just from exposure to language, because words similar in meaning are used in language in similar ways. The successes of…

Abstraction is a core principle of Distributional Semantic Models (DSMs) that learn semantic representations for words by applying dimensional reduction to statistical redundancies in language. Although the posited learning mechanisms vary widely, virtually all DSMs are prototype models in that they create a single abstract representation of a…

This paper explores the concept of dynamically generating definitions using a deep-learning model. We do this by creating a dataset that contains definition entries and contexts associated with each definition. We then fine-tune a GPT-2 based model on the dataset to allow the model to generate contextual definitions. We evaluate our model with…

This aim of this study is to determine the effect of the TPR method on students' vocabulary learning and the factors affecting the effectiveness of this method by combining the findings obtained from both qualitative and quantitative studies. For this purpose, a primary study with 13 quantitative and 7 qualitative findings was included in this…

This article aims to give a comprehensive guide to planning and designing vocabulary tests which include Identifying the skills to be covered by the test; outlining the course content covered; preparing a table of specifications that shows the skill, content topics and number of questions allocated to each; and preparing the test instructions. The…

In this paper, we investigate the effect of neighbourhood density (ND) on vocabulary size in a computational model of vocabulary development. A word has a high ND if there are many words phonologically similar to it. High ND words are more easily learned by infants of all abilities (e.g. Storkel, 2009; Stokes, 2014). We present a neural network…

Ambridge argues that there is widespread agreement among child language researchers that learners store linguistic abstractions. In this commentary the authors first argue that this assumption is incorrect; anti-representationalist/exemplar views are pervasive in theories of child language. Next, the authors outline what has been learned from this…

Prior research has shown that people can learn many nouns (i.e., word--object mappings) from a short series of ambiguous situations containing multiple words and objects. For successful cross-situational learning, people must approximately track which words and referents co-occur most frequently. This study investigates the effects of allowing…

We evaluate here the performance of four models of cross-situational word learning: two global models, which extract and retain multiple referential alternatives from each word occurrence; and two local models, which extract just a single referent from each occurrence. One of these local models, dubbed "Pursuit," uses an associative…

Are associative memories forgotten more quickly than item memories, and does the level of original learning differentially influence forgetting rates? In this study, we addressed these questions by having participants learn single words and word pairs once (Experiment 1), three times (Experiment 2), and six times (Experiment 3) in a massed…

This dissertation presents various machine learning applications for predicting different cognitive states of students while they are using a vocabulary tutoring system, DSCoVAR. We conduct four studies, each of which includes a comprehensive analysis of behavioral and linguistic data and provides data-driven evidence for designing personalized…

Personalized learning has become an important and powerful paradigm catering for various needs, styles, preferences, and modes of learning. Several methods including task recommendations and path planning have recently emerged to effectively implement personalized learning using e-learning systems. The literature shows that the use of task…

Purpose: We investigated four theoretically based latent variable models of word learning in young school-age children. Method: One hundred sixty-seven English-speaking second graders with typical development from three U.S. states participated. They completed five different tasks designed to assess children's creation, storage, retrieval, and…

Computational models have shown that purely statistical knowledge about words' linguistic contexts is sufficient to learn many properties of words, including syntactic and semantic category. For example, models can infer that "postman" and "mailman" are semantically similar because they have quantitatively similar patterns of…

Purpose: Most educational assessments tend to be constructed in a close-ended format, which is easier to score consistently and more affordable. However, recent work has leveraged computation text methods from the information sciences to make open-ended measurement more effective and reliable for older students. This study asks whether such text…