One Step Further

In this qualitative research study, a bidirectional ARS was integrated into a lecture. Students’ perceptions were explored by focusing on their preferences on different question and feedback types, sharing of posts, nickname use, problems, and design suggestions. A total of 25 students participated in focus group interviews. The results showed that students liked the multiple-choice type questions due to the easy answer characteristics, and they found it difficult to text their responses for open-ended questions. The majority of the students preferred getting feedback immediately after asking a question. Students also stated that using ARS ease their shyness. The findings can significantly contribute for understanding the potential of an ARS supporting two-way communication during a lecture-based approach of instruction, also demonstrate that thinking level of the questions with the feasibility of ARS should be investigated together, and the different preferences of students on the question type, feedback type, and nickname use highlight the importance of student characteristics.

Question type and answer related keywords aware question generation

The traditional end-to-end Neural Question Generation (NQG) models tend to generate generic and bland questions, as there are two obscure points: 1) the modifications of the answer in the context can be used as the clues to the answer mentioned in the question, while they are generally not unique and can be used independently for generating diverse questions; 2) the same question content can also be asked in diverse ways, which depends on personal preference in practice. The above-mentioned two points are indeed two variables to conduct question generation, but they are not annotated in the original dataset and are thus ignored by the traditional end-to-end models. In this paper we propose a framework that clarifies those two points through two sub-modules to better conduct question generation. We take experiments based on the GPT-2 model and the SQuAD dataset, and prove that our framework can improve the performance measured by similarity metrics, while it also provides appropriate alternatives for controllable diversity enhancement.

Why Not Interview?

Honesty-Humility is a valuable predictor in personnel selection; however, problems with self-report measures create a need for new tools to judge this trait. Therefore, this research examines the interview as an alternative for assessing Honesty-Humility and how to improve judgments of Honesty-Humility in the interview. Using trait activation theory, we examined the impact of interview question type on Honesty-Humility judgment accuracy. We hypothesized that general personality-tailored questions and probes would increase the accuracy of Honesty-Humility judgments. Nine hundred thirty-three Amazon Mechanical Turk workers watched and rated five interviews. Results found that general questions with probes and specific questions without probes led to the best Honesty-Humility judgments. These findings support the realistic accuracy model and provide implications for Honesty-Humility-based interviews.

How Question Characteristics Impact Answer Outcome on Social Question-and-Answer Websites

Inducing more and higher-quality answers to questions is essential to sustainable development of Social Question-and-Answer (SQA) websites. Previous research has studied factors affecting question success and user motivation in answering questions, but how a question’s own characteristics affect the question’s answer outcome on SQA websites remains unknown. This study examines the impact of the characteristics of a question, namely readability, emotionality, additional descriptions, and question type, on the question’s answer outcome as measured by number of answers, average answer length, and number of “likes” received by answers to the question. Regression analyses reveal that readability, additional descriptions, and question type have significant impact on multiple measurements of answer outcome, while emotionality only affects the average answer length. This study provides insights to SQA website builders as they instruct users on question construction. It also provides insights to SQA website users on how to induce more and higher-quality answers to their questions.

How Question Characteristics Impact Answer Outcomes on Social Question-and-Answer Websites

Inducing more and higher-quality answers to questions is essential to sustainable development of Social Question-and-Answer (SQA) websites. Previous research has studied factors affecting question success and user motivation in answering questions, but how a question’s own characteristics affect the question’s answer outcome on SQA websites remains unknown. This study examines the impact of the characteristics of a question, namely readability, emotionality, additional descriptions, and question type, on the question’s answer outcome as measured by number of answers, average answer length, and number of “likes” received by answers to the question. Regression analyses reveal that readability, additional descriptions, and question type have significant impact on multiple measurements of answer outcome, while emotionality only affects the average answer length. This study provides insights to SQA website builders as they instruct users on question construction. It also provides insights to SQA website users on how to induce more and higher-quality answers to their questions.

Thematic role assignment in non-default verb classes: A cross-linguistic comparison of English and German

The influence of sentential cues (such as animacy and word order) on thematic role interpretation differs as a function of language (MacWhinney et al. 1984). However, existing cross-linguistic research has typically focused on transitive sentences involving agents, and interpretation of non-default verb classes is less well understood. Here, we compared the way in which English and German native speakers – languages known to differ in the cue prominence of animacy and word order – assign thematic roles. We compared their interpretation of sentences containing either default (agent-subject) or non-default (experiencer-subject) verb classes. Animacy of the two noun phrases in a sentence was either animate-inanimate and plausible (e.g. “The men will devour the meals...”) or inanimate-animate and implausible in English (e.g. “The meals will devour the men…”). We examined role assignment by probing for either the actor or undergoer of the sentence. Mixed effects modelling revealed that role assignment was significantly influenced by noun animacy, verb class, question type, and language. Results are interpreted within the Competition Model framework (Bates et al. 1982; MacWhinney et al. 1984), and show that English speakers predominantly relied on word order for thematic role assignment. German speakers relied on word order to a comparatively lesser degree, with animacy a prominent cue. Non-default verbs (experiencer-subject) promoted a non-default comprehension strategy regarding the prominence of sentential cues, particularly in German. Intriguingly, responses were modulated by the probe task, with undergoer probes promoting object-initial interpretations, particularly for German speakers. This suggests that task focus may retroactively influence sentence interpretation.

Do Teacher Questions Function as Assessment for Learning?

This study wants to challenge the robust idea of previous findings revealing that employing a particular question type would necessarily functions as Assessment for Learning (AfL). Besides, this study extends previous research focusing on typology and examines the syntactical forms of questioning in its practice. To gather data, six Indonesian English teachers were observed and audio- recorded, thus, transcribed and analysed following the principle of Conversation Analysis (CA). Except referential type functioning as a teaching technique and a discourse marker choice to discursively extend the classroom talk, the result of analysis corroborates previous studies in that they provide diagnostic information from which a better further action was taken place as highlighted in the AfL. Yet, this might occur as questioning types are syntactically constructed following classroom discourse moves. Thus, the examination of questionings functioning as Assessment for Learning (AfL), aside from types, the syntactical form and classroom discourse moves are important to cope with.

Interaction In the English Language Classroom: A Review of Teacher Questions

Teachers' questions in the English language classroom are an essential way of teaching English. Teachers must be aware of the types of questions that can help students learn the target language. Students, on the other hand, did not actively participate in learning, especially when responding to questions from teachers. To address this issue, teachers must modify their questions using various techniques in order to elicit responses from students. This study examines twenty studies selected from the Google scholar on the role of teacher questioning pattern in motivating students' participation in English language classrooms, as well as the classification of modification questions used by teachers when teaching English. We examine current teachers' questioning patterns before reviewing previous research on the most common question type activity in the English classroom. According to the findings, display questions are frequently used by English language teachers as a better approach than referential and other questioning types. Other knowledge-based analyses were discovered to be carried out in order to extract useful features that reduce the risk of better activity, demonstrating that students continue to struggle with high-dimensional and important subjects when answering referential questions. Finally, we highlight some outstanding issues for future research in this area that researchers should consider.

AdaVQA: Overcoming Language Priors with Adapted Margin Cosine Loss

A number of studies point out that current Visual Question Answering (VQA) models are severely affected by the language prior problem, which refers to blindly making predictions based on the language shortcut. Some efforts have been devoted to overcoming this issue with delicate models. However, there is no research to address it from the view of the answer feature space learning, despite the fact that existing VQA methods all cast VQA as a classification task. Inspired by this, in this work, we attempt to tackle the language prior problem from the viewpoint of the feature space learning. An adapted margin cosine loss is designed to discriminate the frequent and the sparse answer feature space under each question type properly. In this way, the limited patterns within the language modality can be largely reduced to eliminate the language priors. We apply this loss function to several baseline models and evaluate its effectiveness on two VQA-CP benchmarks. Experimental results demonstrate that our proposed adapted margin cosine loss can enhance the baseline models with an absolute performance gain of 15\% on average, strongly verifying the potential of tackling the language prior problem in VQA from the angle of the answer feature space learning.