Task analysis

About: Task analysis is a research topic. Over the lifetime, 10432 publications have been published within this topic receiving 283481 citations.

Knowledge Transfer from Weakly Labeled Audio Using Convolutional Neural Network for Sound Events and Scenes

Abstract: In this work we propose approaches to effectively transfer knowledge from weakly labeled web audio data. We first describe a convolutional neural network (CNN) based framework for sound event detection and classification using weakly labeled audio data. Our model trains efficiently from audios of variable lengths; hence, it is well suited for transfer learning. We then propose methods to learn representations using this model which can be effectively used for solving the target task. We study both transductive and inductive transfer learning tasks, showing the effectiveness of our methods for both domain and task adaptation. We show that the learned representations using the proposed CNN model generalizes well enough to reach human level accuracy on ESC-50 sound events dataset and sets state of art results on this dataset. We further use them for acoustic scene classification task and once again show that our proposed approaches suit well for this task as well. We also show that our methods are helpful in capturing semantic meanings and relations as well. Moreover, in this process we also set state-of-art results on Audioset dataset using balanced training set.

Enhancing Learners' Communication Skills through Synchronous Electronic Interaction and Task‐Based Instruction

Abstract: Online interactive exchange offers the learner many opportunities to use the target language to negotiate both meaning and form in a social context that is crucial for second language acquisition. This paper discusses a pilot study using synchronous electronic chats combined with task-based instruction (TBI) to enhance learners' communication skills. TBI focuses on the two-way exchange of information on real-life topics. This pilot study shows that computer-mediated communication using less structure-controlled but more open-ended exchange had a significant impact on the process of language learning. Students benefited from online task-based activities because they had to access different functional skills to construct and negotiate meaning collaboratively. However, foreign language educators need to be aware that the quick cyberspace interactions impeded students from producing correct and coherent discourse, especially during learner-learner interaction. One corrective technique is to make students reexamine and revise their exchanges with guided instruction.

Learner acquaintanceship and oral proficiency test pair-task performance:

Abstract: This paper explores the effect on pair-task performance of test-takers’ familiarity with their partner. O’Sullivan (2000a), like McNamara (1996), sees test performance as being affected by a number of factors, related to the test-taker, the interlocutor and the task. Evidence from the psychology literature, and anecdotal evidence from language learners and teachers, suggests that familiarity with one’s partner in an interactive task might positively affect performance, although in the only published study to date, Porter (1991a) found no evidence to support this hypothesis.A group of 32 Japanese learners performed a series of three tasks (personal information exchange, narrative, and decision making), once with a friend, and again with a person who was not known to them. All performances were video-recorded, awarded scores by trained raters and transcribed for analysis. Results suggest evidence of an `acquaintanceship’ effect, with subjects achieving higher scores when working with a friend. However, anal...

Learning L2 German vocabulary through reading: the effect of three enhancement techniques compared

Abstract: This study investigated three techniques designed to increase the chances that second language (L2) readers look up and learn unfamiliar words during and after reading an L2 text. Participants in the study, 137 college students in Belgium (L1 = Dutch, L2 = German), were randomly assigned to one of four conditions, forming combinations of two between-subject factors: ± prereading test announcement and ± postreading vocabulary task. Comprehension questions were used to direct participants' attention to half of the 16 target words in this study, creating the within-subject factor (word relevance). Participants accomplished the experimental tasks at computers. They could look up the meaning of unfamiliar words in an online dictionary. The dependent variables are the following: use of online dictionary during reading, performance on a word-form recognition test, and performance on two word-meaning recall tests (immediate and delayed). Test announcement and word relevance substantially prompted participants to use the online dictionary more. Only test announcement and vocabulary task (not word relevance) affected performance in the word recognition test positively. Both word relevance and postreading vocabulary task substantially affected word retention in the recall posttests. These findings, together with those of the studies reviewed, provide robust evidence that the low incidence of vocabulary acquisition through reading ("input only") can be substantially boosted by techniques that make students look up the meaning of unknown words, process their form-meaning relationship elaborately, and process them again after reading ("input plus").