Search

OpenRefine is a popular open-source data cleaning tool. It allows users to export a previously executed data cleaning workflow in a JSON format for possible reuse on other datasets. We have developed or2yw, a novel tool that maps a JSON-formatted OpenRefine operation history to a YesWorkflow (YW) model, which then can be visualized and queried using the YW tool. The latter was originally developed to allow researchers a simple way to annotate their program scripts in order to reveal the workflow steps and dataflow dependencies implicit in those scripts. With or2yw the user can automatically generate YW models from OpenRefine operation histories, thus providing a 'workflow view' on a previously executed sequence of data cleaning operations. The or2yw tool can generate different types of YesWorkflow models, e.g., a linear model which mirrors the sequential execution order of operations in OpenRefine, and a \emph{parallel model} which reveals independent workflow branches, based on a simple analysis of dependencies between steps: if two operations are independent of each other (e.g., when the columns they read and write do not overlap) then these can be viewed as parallel steps in the data cleaning workflow. The resulting YW models can be understood as a form of prospective provenance, i.e., knowledge artifacts that can be queried and visualized (i) to help authors document their own data cleaning workflows, thereby increasing transparency, and (ii) to help other users, who might want to reuse such workflows, to understand them better.

Automated knowledge discovery from trending chemical literature is essential for more efficient biomedical research. How to extract detailed knowledge about chemical reactions from the core chemistry literature is a new emerging challenge that has not been well studied. In this paper, we study the new problem of fine-grained chemical entity typing, which poses interesting new challenges especially because of the complex name mentions frequently occurring in chemistry literature and graphic representation of entities. We introduce a new benchmark data set (CHEMET) to facilitate the study of the new task and propose a novel multi-modal representation learning framework to solve the problem of fine-grained chemical entity typing by leveraging external resources with chemical structures and using cross-modal attention to learn effective representation of text in the chemistry domain. Experiment results show that the proposed framework outperforms multiple state-of-the-art methods.

The quadratic computational and memory complexities of large Transformers have limited their scalability for long document summarization. In this paper, we propose Hepos, a novel efficient encoder-decoder attention with head-wise positional strides to effectively pinpoint salient information from the source. We further conduct a systematic study of existing efficient self-attentions. Combined with Hepos, we are able to process ten times more tokens than existing models that use full attentions. For evaluation, we present a new dataset, GovReport, with significantly longer documents and summaries. Results show that our models produce significantly higher ROUGE scores than competitive comparisons, including new state-of-the-art results on PubMed. Human evaluation also shows that our models generate more informative summaries with fewer unfaithful errors., Comment: Accepted at NAACL 2021 as a long paper

To combat COVID-19, both clinicians and scientists need to digest vast amounts of relevant biomedical knowledge in scientific literature to understand the disease mechanism and related biological functions. We have developed a novel and comprehensive knowledge discovery framework, COVID-KG to extract fine-grained multimedia knowledge elements (entities and their visual chemical structures, relations, and events) from scientific literature. We then exploit the constructed multimedia knowledge graphs (KGs) for question answering and report generation, using drug repurposing as a case study. Our framework also provides detailed contextual sentences, subfigures, and knowledge subgraphs as evidence., Comment: 12 pages, Accepted by Proceedings of 2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics System Demonstrations, for resources see http://blender.cs.illinois.edu/covid19/, for video see http://159.89.180.81/demo/covid/Covid-KG_DemoVideo.mp4, for slides see https://eaglew.github.io/files/Covid-KG_DemoVideo_with_ethics.pdf