Mitigating Biases in CORD-19 for Analyzing COVID-19 Literature.

On the behest of the Office of Science and Technology Policy in the White House, six institutions, including ours, have created an open research dataset called COVID-19 Research Dataset (CORD-19) to facilitate the development of question-answering systems that can assist researchers in finding relevant research on COVID-19. As of May 27, 2020, CORD-19 includes more than 100,000 open access publications from major publishers and PubMed as well as preprint articles deposited into medRxiv, bioRxiv, and arXiv. Recent years, however, have also seen question-answering and other machine learning systems exhibit harmful behaviors to humans due to biases in the training data. It is imperative and only ethical for modern scientists to be vigilant in inspecting and be prepared to mitigate the potential biases when working with any datasets. This article describes a framework to examine biases in scientific document collections like CORD-19 by comparing their properties with those derived from the citation behaviors of the entire scientific community. In total, three expanded sets are created for the analyses: 1) the enclosure set CORD-19E composed of CORD-19 articles and their references and citations, mirroring the methodology used in the renowned "A Century of Physics" analysis; 2) the full closure graph CORD-19C that recursively includes references starting with CORD-19; and 3) the inflection closure CORD-19I, that is, a much smaller subset of CORD-19C but already appropriate for statistical analysis based on the theory of the scale-free nature of the citation network. Taken together, all these expanded datasets show much smoother trends when used to analyze global COVID-19 research. The results suggest that while CORD-19 exhibits a strong tilt toward recent and topically focused articles, the knowledge being explored to attack the pandemic encompasses a much longer time span and is very interdisciplinary. A question-answering system with such expanded scope of knowledge may perform better in understanding the literature and answering related questions. However, while CORD-19 appears to have topical coverage biases compared to the expanded sets, the collaboration patterns, especially in terms of team sizes and geographical distributions, are captured very well already in CORD-19 as the raw statistics and trends agree with those from larger datasets.

Robust Surface-Engineered Tape-Cast and Extrusion Methods to Fabricate Electrically-Conductive Poly(vinylidene fluoride)/Carbon Nanotube Filaments for Corrosion-Resistant 3D Printing Applications.

We developed a poly(vinylidene fluoride)/carbon nanotube (PVDF-MWCNT) filament as a feed for printing of electrically-conductive and corrosion-resistant functional material by fused filament fabrication (FFF). Using an environment-friendly procedure to fabricate PVDF-MWCNT filament, we achieved the best reported electrical conductivity of printable PVDF-MWCNT filament of 28.5 S cm-1 (90 wt% PVDF and 10 wt% CNT). The PVDF-MWCNT filaments are chemically stable in acid, base, and salt solution, with no significant changes in electrical conductivity and mass of the filaments. Our processing method is robust and allow a uniform mixture of PVDF and CNT with a wide range of CNT percentage up to 99.9%. We demonstrated the printing of PVDF-MWCNT filaments to create 3D shapes; printed using a low-cost commercial consumer-grade FFF 3D printer. We found many adjustments of printer parameters are needed to print filament with CNT content >10 wt%, but easier printing for CNT content ≤10 wt%. Since this was due to printer limitation, we believed that PVDF-MWCNT with higher CNT percentage (to a certain limit) and larger electrical conductivity could be printed with a custom-built printer (for example stronger motor). PVDF-MWCNT filament shows higher electrical conductivity (28.5 S cm-1) than compressed composite (8.8 S cm-1) of the same 10 wt% of CNT, due to more alignment of CNT in the longitudinal direction of the extruded filament. Printable PVDF-MWCNT-Fe2O3 (with a functional additive of Fe2O3) showed higher electrical conductivity in the longitudinal direction at the filament core (42 S cm-1) compared to that in the longitudinal direction at the filament shell (0.43 S cm-1) for sample with composition of 60 wt% PVDF, 20 wt% CNT, and 20 wt% Fe2O3, due to extrusion skin effect with segregation of electrically insulating Fe2O3 at the shell surface of PVDF-MWCNT-Fe2O3.

Nanoscopic and Macro-Porous Carbon Nano-foam Electrodes with Improved Mass Transport for Vanadium Redox Flow Batteries.

Although free-standing sheets of multiwalled carbon nanotubes (MWCNT) can provide interesting electrochemical and physical properties as electrodes for redox flow batteries, the full potential of this class of materials has not been accessible as of yet. The conventional fabrication methods produce sheets with micro-porous and meso-porous structures, which significantly resist mass transport of the electrolyte during high-current flow-cell operation. Herein, we developed a method to fabricate high performance macro-porous carbon nano-foam free standing sheets (Puffy Fibers, PF), by implementing a freeze-drying step into our low cost and scalable surface-engineered tape-casting (SETC) fabrication method, and we show the improvement in the performance attained as compared with a MWCNT sheet lacking any macro pores (Tape-cast, TC). We attribute the higher performance attained by our in-lab fabricated PF papers to the presence of macro pores which provided channels that acted as pathways for electrolytic transport within the bulk of the electrode. Moreover, we propose an electrolytic transport mechanism to relate ion diffusivity to different pore sizes to explain the different modes of charge transfer in the negative and the positive electrolytes. Overall, the PF papers had a high wettability, high porosity, and a large surface area, resulting in improved electrochemical and flow-cell performances.

A Review of Microsoft Academic Services for Science of Science Studies.

Since the relaunch of Microsoft Academic Services (MAS) 4 years ago, scholarly communications have undergone dramatic changes: more ideas are being exchanged online, more authors are sharing their data, and more software tools used to make discoveries and reproduce the results are being distributed openly. The sheer amount of information available is overwhelming for individual humans to keep up and digest. In the meantime, artificial intelligence (AI) technologies have made great strides and the cost of computing has plummeted to the extent that it has become practical to employ intelligent agents to comprehensively collect and analyze scholarly communications. MAS is one such effort and this paper describes its recent progresses since the last disclosure. As there are plenty of independent studies affirming the effectiveness of MAS, this paper focuses on the use of three key AI technologies that underlies its prowess in capturing scholarly communications with adequate quality and broad coverage: (1) natural language understanding in extracting factoids from individual articles at the web scale, (2) knowledge assisted inference and reasoning in assembling the factoids into a knowledge graph, and (3) a reinforcement learning approach to assessing scholarly importance for entities participating in scholarly communications, called the saliency, that serves both as an analytic and a predictive metric in MAS. These elements enhance the capabilities of MAS in supporting the studies of science of science based on the GOTO principle, i.e., good and open data with transparent and objective methodologies. The current direction of development and how to access the regularly updated data and tools from MAS, including the knowledge graph, a REST API and a website, are also described.