首页|University of Leipzig Researcher Discusses Research in Machine Learning (Is medi eval distant viewing possible? : Extending and enriching annotation of legacy im age collections using visual analytics)
University of Leipzig Researcher Discusses Research in Machine Learning (Is medi eval distant viewing possible? : Extending and enriching annotation of legacy im age collections using visual analytics)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News - Investigators discuss new findings in artificial intelligence. According to news originating from Leipzig, Germany, by NewsRx correspondents, research stated, “Distant viewing approaches have typica lly used image datasets close to the contemporary image data used to train machi ne learning models.” The news editors obtained a quote from the research from University of Leipzig: “To work with images from other historical periods requires expert annotated dat a, and the quality of labels is crucial for the quality of results. Especially w hen working with cultural heritage collections that contain myriad uncertainties , annotating data, or re-annotating, legacy data is an arduous task. In this pap er, we describe working with two pre-annotated sets of medieval manuscript image s that exhibit conflicting and overlapping metadata. Since a manual reconciliati on of the two legacy ontologies would be very expensive, we aim (1) to create a more uniform set of descriptive labels to serve as a “bridge” in the combined da taset, and (2) to establish a high-quality hierarchical classification that can be used as a valuable input for subsequent supervised machine learning. To achie ve these goals, we developed visualization and interaction mechanisms, enabling medievalists to combine, regularize and extend the vocabulary used to describe t hese, and other cognate, image datasets.”
University of LeipzigLeipzigGermanyEuropeCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
