Co-similar malware infection patterns as a predictor of future risk.

The internet is flooded with malicious content that can come in various forms and lead to information theft and monetary losses. From the ISP to the browser itself, many security systems act to defend the user from such content. However, most systems have at least one of three major limitations: 1) they are not personalized and do not account for the differences between users, 2) their defense mechanism is reactive and unable to predict upcoming attacks, and 3) they extensively track and use the user's activity, thereby invading her privacy in the process. We developed a methodological framework to predict future exposure to malicious content. Our framework accounts for three factors-the user's previous exposure history, her co-similarity to other users based on their previous exposures in a conceptual network, and how the network evolves. Utilizing over 20,000 users' browsing data, our approach succeeds in achieving accurate results on the infection-prone portion of the population, surpassing common methods, and doing so with as little as 1/1000 of the personal information it requires.

Predicting and improving patient-level antibiotic adherence.

Low adherence to prescribed medications causes substantial health and economic burden. We analyzed primary data from electronic medical records of 250,000 random patients from Israel's Maccabi Healthcare services from 2007 to 2017 to predict whether a patient will purchase a prescribed antibiotic. We developed a decision model to evaluate whether an intervention to improve purchasing adherence is warranted for the patient, considering the cost of the intervention and the cost of non-adherence. The best performing prediction model achieved an average area under the receiver operating characteristic curve (AUC) of 0.684, with 82% accuracy in detecting individuals who had less than 50% chance of purchasing a prescribed drug. Using the decision model, an adherence intervention targeted to patients whose predicted purchasing probability is below a specified threshold can increase the number of prescriptions filled while generating significant savings compared to no intervention - on the order of 6.4% savings and 4.0% more prescriptions filled for our dataset. We conclude that analysis of large-scale patient data from electronic medical records can help predict the probability that a patient will purchase a prescribed antibiotic and can provide real-time predictions to physicians, who can then counsel the patient about medication importance. More broadly, in-depth analysis of patient-level data can help shape the next generation of personalized interventions.