Forbus, Kenneth D. (Northwestern University) | Garnier, Bridget (University of Wisconsin-Madison) | Tikoff, Basil (University of Wisconsin-Madison) | Marko, Wayne (Northwestern University) | Usher, Madeline (Northwestern University) | McLure, Matthew (Northwestern University)
Sketching can be a valuable tool for science education, but it is currently underutilized. Sketch worksheets were developed to help change this, by using AI technology to give students immediate feedback and to give instructors assistance in grading. Sketch worksheets use visual representations automatically computed by CogSketch, which are combined with conceptual information from the OpenCyc ontology. Feedback is provided to students by comparing an instructor’s sketch to a student’s sketch, using the Structure-Mapping Engine. This paper describes our experiences in deploying sketch worksheets in two types of classes: Geoscience and AI. Sketch worksheets for introductory geoscience classes were developed by geoscientists at University of Wisconsin-Madison, authored using CogSketch and used in classes at both Wisconsin and Northwestern University. Sketch worksheets were also developed and deployed for a knowledge representation and reasoning course at Northwestern. Our experience indicates that sketch worksheets can provide helpful on-the-spot feedback to students, and significantly improve grading efficiency, to the point where sketching assignments can be more practical to use broadly in STEM education.
Geoscience has traditionally struggled with metadata for datasets. There are many fixed metadata schemas to facilitate the search and retrieval of datasets but they have trouble representing the richness of datasets coming from an ever-evolving science. For the IRI Data Library of datasets, we want a more relational and extensible paradigm for metadata. The RDF/OWL framework in the form of OWL ontologies presents that possibility.
The interleaving of human, machine, and semantics have the potential to overcome some of the issues currently surrounding Big Data. Semantic technologies, in particular, have been shown to adequately address data integration when dealing with data size, variety, and complexity of data sources – the very definition of Big Data. Yet, for some tasks, semantic algorithms do not reach a level of accuracy that many production environments require. In this position paper, we argue that augmenting such algorithms with crowdsourcing is a viable solution. In particular, we examine Big Data within the geosciences and describe outstanding questions regarding the merger of crowdsourcing and semantics. We present our ongoing work in this area and discuss directions for future research.