Learning to rank for<Emphasis Type="Italic"> why</Emphasis>-question answering     

Suzan Verberne  Hans van Halteren  Daphne Theijssen  Stephan Raaijmakers  Lou Boves 《Information Retrieval》2011,14(2):107-132

In this paper, we evaluate a number of machine learning techniques for the task of ranking answers to why-questions. We use TF-IDF together with a set of 36 linguistically motivated features that characterize questions and answers. We experiment with a number of machine learning techniques (among which several classifiers and regression techniques, Ranking SVM and SVM ^map ) in various settings. The purpose of the experiments is to assess how the different machine learning approaches can cope with our highly imbalanced binary relevance data, with and without hyperparameter tuning. We find that with all machine learning techniques, we can obtain an MRR score that is significantly above the TF-IDF baseline of 0.25 and not significantly lower than the best score of 0.35. We provide an in-depth analysis of the effect of data imbalance and hyperparameter tuning, and we relate our findings to previous research on learning to rank for Information Retrieval.  相似文献   

Effects of self-assessment feedback on self-assessment and task-selection accuracy     

Raaijmakers  Steven F.  Baars  Martine  Paas  Fred  van Merriënboer  Jeroen J. G.  van Gog  Tamara 《Metacognition and Learning》2019,14(1):21-42

Metacognition and Learning - Effective self-regulated learning in settings in which students can decide what tasks to work on, requires accurate self-assessment (i.e., a judgment of own level of...  相似文献   

Training self-regulated learning skills with video modeling examples: Do task-selection skills transfer?     

Steven F. Raaijmakers  Martine Baars  Lydia Schaap  Fred Paas  Jeroen van Merriënboer  Tamara van Gog 《Instructional Science》2018,46(2):273-290

Self-assessment and task-selection skills are crucial in self-regulated learning situations in which students can choose their own tasks. Prior research suggested that training with video modeling examples, in which another person (the model) demonstrates and explains the cyclical process of problem-solving task performance, self-assessment, and task-selection, is effective for improving adolescents’ problem-solving posttest performance after self-regulated learning. In these examples, the models used a specific task-selection algorithm in which perceived mental effort and self-assessed performance scores were combined to determine the complexity and support level of the next task, selected from a task database. In the present study we aimed to replicate prior findings and to investigate whether transfer of task-selection skills would be facilitated even more by a more general, heuristic task-selection training than the task-specific algorithm. Transfer of task-selection skills was assessed by having students select a new task in another domain for a fictitious peer student. Results showed that both heuristic and algorithmic training of self-assessment and task-selection skills improved problem-solving posttest performance after a self-regulated learning phase, as well as transfer of task-selection skills. Heuristic training was not more effective for transfer than algorithmic training. These findings show that example-based self-assessment and task-selection training can be an effective and relatively easy to implement method for improving students’ self-regulated learning outcomes. Importantly, our data suggest that the effect on task-selection skills may transfer beyond the trained tasks, although future research should establish whether this also applies when trained students perform novel tasks themselves.  相似文献