Science4Data Article Insights

https://ai.googleblog.com/2020/07/exploring-faster-screening-with-fewer.html

Exploring Faster Screening with Fewer Tests via Bayesian Group Testing

2020-07-14 00:00:00, Posted by Marco Cuturi and Jean-Philippe Vert, Research Scientists, Google Research, Brain Team, Google AI Blog

Content Categorization
/Science

Word Count:
1909

Words/Sentence:
31

Reading Time:
12.73 min

Reading Quality:
Adept

Readability:
13th to 15th

Media Sentiment
Proprietary sentiment analysis on both the headline and body text of the article. Sentiment scores range from -1 (very negative sentiment) to 1 (very positive sentiment).
RCS Analysis
Relative scoring for Risk, Crisis, and Security language within the article.
Risk Score
Scoring based on the composite risk, security and crisis language within an article compared to a baseline of historic analysis across thousands of diverse articles.
PESTEL Scope
Analysis of article orientation across the PESTEL macro-environmental analysis framework. Learn more about PESTEL.
Entity Word Cloud
Key people, places, organizations and events referenced in the article, weighted by frequency and colored based on contextual sentiment.
Auto Summary
Condensing key features of the article based on salience analysis. Helpful for “gisting” the article in a time crunch.

This perspective allows us to seamlessly incorporate additional prior knowledge on infection, such as when we suspect some individuals to be more likely than others to carry the pathogen, based for instance on contact tracing data or answers to a questionnaire.

Our first contribution is to adopt a probabilistic perspective, and form thousands of plausible hypotheses of infection distributions given test outcomes, rather than trust test results to be 100% reliable as Dorfman did.

Our second contribution is to propose algorithms that can take advantage of these hypotheses to form new groups, and therefore direct the gathering of new evidence, to narrow down as quickly as possible to the "true" infection hypothesis, and close the case with as little testing effort as possible.

Finally, some strategies are adaptive, proposing groups based on test results already observed (including Dorfman's, since it proposes to re-test individuals that appeared in positive groups), whereas others stick to a non-adaptive setting in which groups are known beforehand or drawn at random.

Intuitively, if k=1 and one can only propose a single new group to test, there would be clear advantage in building that group such that its test outcome is as uncertain as possible, i.e., with a probability that it returns positive as close to 50% as possible, given the current set of hypotheses.

Keywords

Interested in Learning More?

Article Analysis is at the foundation of powerful media monitoring and insights. Learn what you can build with powerful curated search engines, real-time listening and trend analysis on the topics, markets and companies critical to your organization.