IBM today announced the launch of services intended to furnish researchers with resources to fight the novel coronavirus. The company made molecules identified by AI as therapeutic candidates available under an open license, and it introduced a free version of its Functional Genomics Platform to support genome features discovery. Additionally, it provided free access to over 1,000 pieces of evidence-based curated COVID-19 and infectious disease content, and it rolled out an AI search engine trained on the COVID-19 Open Research Dataset to allow researchers to quickly find answers to questions.
According to Dario Gil, director of IBM Research, IBM created a new AI-generative framework that can rapidly create novel peptides, proteins, drug candidates, and materials, which it applied against three targets to create 3,000 new small molecules as potential COVID-19 therapeutic candidates. Researchers can study them via an interactive tool to understand their characteristics and relationship to COVID-19, and to identify molecules that might have desirable properties to be pursued in drug development.
“The traditional drug discovery pipeline relies on a library of compounds that are screened, improved, and tested to determine safety and efficacy,” said Gil. “In dealing with new pathogens such as SARS-CoV-2, there is the potential to enhance the compound libraries with additional novel compounds.”
The Functional Genomics Platform complements this with a cloud-based repository of genes, proteins, and other molecular targets from sequenced viral and bacterial organisms, with connections precomputed to help speed up the discovery of molecular targets required for drug design, test development, and treatment. Researchers working on COVID-19 can now request access to the interface; previously, it was only available to government agencies, academic institutions, and other organizations for bacterial study.
In a related development, starting today, clinicians and health care workers will gain access to COVID-19 content from IBM: Micromedex and EBSCO DynaMed, two holistic drug information and medical decision support platforms. (Micromedex is used by more than 4,500 hospitals and health systems worldwide, while DynaMed provides peer-reviewed clinical content including literature reviews in 28 specialties.) They’ll be able to take advantage of state-of-the-art drug and disease data and provide patients with educational handouts containing actionable information.
Lastly, IBM says it trained an AI search system on the COVID-19 Open Research Dataset, which contains thousands of scientific papers prepared by the White House and a coalition of research groups, in addition to licensed databases from DrugBank and GenBank. The freely available tool extracts embedded text, tables, and graphics to answer specific queries from vetted researchers.
Watson Assistant for Citizens’ debut comes after IBM made available a map on The Weather Channel to track the spread of COVID-19, mainly using data from governments as well as the World Health Organization. The company also built a dashboard on top of its Cognos Analytics suite that’s designed to help researchers, data scientists, and media analyze and filter coronavirus information down to the county level.
IBM this week debuted a chatbot that answers coronavirus questions by phone or text, and last week, the company announced it would coordinate an effort to make supercomputing capacity available to researchers in order to help identify treatments, viable mitigation strategies, and vaccines for COVID-19. IBM also launched a new Call for Code Global Challenge that will encourage developers to build open source technologies that address several areas, including crisis communication during an emergency, ways to improve remote learning, and how to inspire cooperative local communities.
AI tools could improve fake news detection by analyzing users’ interactions and comments
In a paper published on the preprint server Arxiv.org, researchers affiliated with Microsoft and Arizona State University propose an approach to detecting fake news that leverages a technique called weak social supervision. They say that by enabling the training of fake news-detecting AI even in scenarios where labeled examples aren’t available, weak social supervision opens the door to exploring how aspects of user interactions indicate news might be misleading.
According to the Pew Research Center, approximately 68% of U.S. adults got their news from social media in 2018 — which is worrisome considering misinformation about the pandemic continues to go viral, for instance. Companies from Facebook and Twitter to Google are pursuing automated detection solutions, but fake news remains a moving target owing to its topical and stylistic diverseness.
Building on a study published in April, the coauthors of this latest work suggest that weak supervision — where noisy or imprecise sources provide data labeling signals — could improve fake news detection accuracy without requiring fine-tuning. To this end, they built a framework dubbed Tri-relationship for Fake News (TiFN) that models social media users and their connections as an “interaction network” to detect fake news.
Interaction networks describe the relationships among entities like publishers, news pieces, and users; given an interaction network, TiFN’s goal is to embed different types of entities, following from the observation that people tend to interact with like-minded friends. In making its predictions, the framework also accounts for the fact that connected users are more likely to share similar interests in news pieces; that publishers with a high degree of political bias are more likely to publish fake news; and that users with low credibility are more likely to spread fake news.
To test whether TiFN’s weak social supervision could help to detect fake news effectively, the team validated it against a Politifact data set containing 120 true news and 120 verifiably fake pieces shared among 23,865 users. Versus baseline detectors that consider only news content and some social interactions, they report that TiFN achieved between 75% to 87% accuracy even with a limited amount of weak social supervision (within 12 hours after the news was published).
In another experiment involving a separate custom framework called Defend, the researchers sought to use as a weak supervision signal news sentences and user comments explaining why a piece of news is fake. Tested on a second Politifact data set consisting of 145 true news and 270 fake news pieces with 89,999 comments from 68,523 users on Twitter, they say that Defend achieved 90% accuracy.
[W]ith the help of weak social supervision from publisher-bias and user-credibility, the detection performance is better than those without utilizing weak social supervision. We [also] observe that when we eliminate news content component, user comment component, or the co-attention for news contents and user comments, the performances are reduced. [This] indicates capturing the semantic relations between the weak social supervision from user comments and news contents is important,” wrote the researchers. “[W]e can see within a certain range, more weak social supervision leads to a larger performance increase, which shows the benefit of using weak social supervision.”
Let’s block ads! (Why?)
Big Data – VentureBeat