This paper investigates whether the power of the models pre-trained on text data, such as BERT, can be transferred to general token sequence classification applications. To verify pre-trained models' transferability, we test the pre-trained models on
text classification tasks with meanings of tokens mismatches, and real-world non-text token sequence classification data, including amino acid, DNA, and music. We find that even on non-text data, the models pre-trained on text converge faster, perform better than the randomly initialized models, and only slightly worse than the models using task-specific knowledge. We also find that the representations of the text and non-text pre-trained models share non-trivial similarities.
A text retrieval system for language learning returns reading materials at the appropriate difficulty level for the user. The system typically maintains a learner model on the user's vocabulary knowledge, and identifies texts that best fit the model.
As the user's language proficiency increases, model updates are necessary to retrieve texts with the corresponding lexical complexity. We investigate an open learner model that allows user modification of its content, and evaluate its effectiveness with respect to the amount of user update effort. We compare this model with the graded approach, in which the system returns texts at the optimal grade. When the user makes at least half of the expected updates to the open learner model, simulation results show that it outperforms the graded approach in retrieving texts that fit user preference for new-word density.
Automatic personalized corrective feedback can help language learners from different backgrounds better acquire a new language. This paper introduces a learner English dataset in which learner errors are accompanied by information about possible erro
r sources. This dataset contains manually annotated error causes for learner writing errors. These causes tie learner mistakes to structures from their first languages, when the rules in English and in the first language diverge. This new dataset will enable second language acquisition researchers to computationally analyze a large quantity of learner errors that are related to language transfer from the learners' first language. The dataset can also be applied in personalizing grammatical error correction systems according to the learners' first language and in providing feedback that is informed by the cause of an error.
The ultimate goal of this study is to enhance the cultural competence of the Syrian
students at both the school and university phases throughout the teaching of English. The
first part of this study provides what many researchers think about the ro
les of the textbook
and the teacher in strengthening the relationship between the learner and his/her own
culture and/or the other cultures including the target culture. In the second part of this
study, an analysis from a cultural perspective of the Syrian context is provided. This
analysis focuses on the English language textbooks used at the Syrian schools and at
Tishreen university. Then it moves to both the Syrian teachers and leaners. The last part of
this study gives some suggestions that could be helpful in strengthening the cultural
awareness in the Syrian educational system.
The study aimed to identify the effectiveness of the proposed
teaching scouting methods of "problem solving" a model in the
collection of knowledge-based high in the Arabic language through
the study sample, "the control group and the experimental
" and a
comparison between the views of gender on the effectiveness of the
proposed teaching methods and the existing views on the learner's
effort and the teacher in the achievement of the collection cybernetic
high in the Arabic language, So was teaching program design, and
judgment, and designed final tests (before me, and after me / Direct
and deferred) for each lesson of the selected lessons from the Arab
rule my (archives of texts and applaud).
The present research aims to illustrate teachers' use of tasks which
promote learner autonomy in ELT classroom interaction.
Extracts taken from video-recorded classroom interactions have been
selected for analysis and examination in an attempt to
identify the
strategies which give learners more responsibilities for learning.
The findings show that teachers use certain strategies such as
structuring classroom discourse to add support for learners' independent
learning, providing an amount of choice for self-directed learning,
collectively constructing the discourse, setting up collaborative tasks to
be completed independently, and engaging students with problem-solving
activities.
The findings indicate that such strategies train learners to develop
independence in their approach to learning. It is proposed that the
implications for promoting learner autonomy through classroom
interaction should be incorporated into teacher-training programmes.
