Sentence examples for answer selection from inspiring English sources

In this paper, we propose a recurrent convolutional neural network (RCNN) for answer selection in community question answering (CQA).

The results prove the effectiveness of the proposed model on the task of answer selection in CQA.

In some cases, students are also asked to provide written explanations justifying their answer selection (Koretsky et al. 2016).

We removed one item (question 3) from the analysis due to a typo in the answer selection on two of the tests.

By integrating thread-level features into QA matching, our model achieves the best performance of Macro-F1 58.77%, which is 1.6% higher than the best submitted system of the answer selection task in SemEval2015.

While these and our own exploratory studies [25, 21, 22] present interesting results, no comprehensive study has been done so far that looked into the strengths and weaknesses of such systems considering different kinds of users, question types, and answer selection schemes (e.g., open-ended vs. multiple choice).

We also examined misconception uses in their answer selections.

We also looked at misconception uses in their answer selections.

We first ran a line item analysis to look for trends in student responses and to identify item discrimination power as evidence for purposeful answer selections.

The use of misconceptions as distractors in the multiple-choice questions allow for more complete assessment of student conceptual knowledge (Anderson et al. 2002); based on their answer selections, students who used the tutorial showed a decrease in selection of misconceptions from pre to post-test (Fig. 3).

Such answer selections do not reflect the transformation that is the hallmark of conceptual change.