2 results
Graph-based methods for discrete choice
- Kiran Tomlinson, Austin R. Benson
-
- Journal:
- Network Science / Volume 12 / Issue 1 / March 2024
- Published online by Cambridge University Press:
- 06 November 2023, pp. 21-40
-
- Article
-
- You have access Access
- Open access
- HTML
- Export citation
-
Choices made by individuals have widespread impacts—for instance, people choose between political candidates to vote for, between social media posts to share, and between brands to purchase—moreover, data on these choices are increasingly abundant. Discrete choice models are a key tool for learning individual preferences from such data. Additionally, social factors like conformity and contagion influence individual choice. Traditional methods for incorporating these factors into choice models do not account for the entire social network and require hand-crafted features. To overcome these limitations, we use graph learning to study choice in networked contexts. We identify three ways in which graph learning techniques can be used for discrete choice: learning chooser representations, regularizing choice model parameters, and directly constructing predictions from a network. We design methods in each category and test them on real-world choice datasets, including county-level 2016 US election results and Android app installation and usage data. We show that incorporating social network structure can improve the predictions of the standard econometric choice model, the multinomial logit. We provide evidence that app installations are influenced by social context, but we find no such effect on app usage among the same participants, which instead is habit-driven. In the election data, we highlight the additional insights a discrete choice framework provides over classification or regression, the typical approaches. On synthetic data, we demonstrate the sample complexity benefit of using social information in choice models.
Guiding Principles and Practices for Healthcare Outbreak Notification and Disclosures: CORHA Policy Workgroup Framework
- Maureen Tierney, Moon Kim, Christopher Baliga, Martha Ngoh, Kiran Perkins, Marion Kainer, Lisa McGiffert, Richard Martinello, Meredith Allen, Kate Heyer, Lisa Tomlinson, Joseph Perz, Dawn Terashita
-
- Journal:
- Infection Control & Hospital Epidemiology / Volume 41 / Issue S1 / October 2020
- Published online by Cambridge University Press:
- 02 November 2020, pp. s480-s481
- Print publication:
- October 2020
-
- Article
-
- You have access Access
- Export citation
-
Background: Outbreaks of infections in healthcare negatively impact patient outcomes and experience. Transparency is critical to engendering trust and optimizing health. Consistent guidance is not available regarding when to report a possible outbreak of healthcare-associated infections (HAIs) to public health and when to notify a limited population or to publicly disclose the occurrence of HAI. Recent analyses of state public health policies revealed that most states address reporting of outbreaks using terms such as clusters, unusual occurrences, or incidences over baseline. Specific wording about healthcare outbreaks or guidance for notifying patients or public is often absent. Thus, HAI outbreak notification and disclosure guidance and practices vary significantly around the country. A best-practice guidance document will provide clarity for when such reporting should occur. Methods: The Council for Outbreak Response: HAI and Antimicrobial-Resistant Pathogens (CORHA) has undertaken the task of developing this guidance by forming a multidiscipline policy work group with representation from its partner organizations. This work group has been tasked with creating a general framework that will guide notification and disclosure in the context of a possible HAI outbreak. The draft guidance document has been developed over several months of telephone and in-person conferences among work group members. Results: The standardized actions stemming from the guiding principles and recommended practices for conducting step 1 (immediate notification, initial and critical communications that occur when an outbreak is first suspected), were arranged in a table format with rows representing stakeholders and constituents to be notified and columns demonstrating the actions to be taken (Fig. 1). As an investigation progresses, notification should be revisited, especially if an investigation’s scope expands. The principles and practices for step 2 (expanded notification) have also been drafted in a table format. Next, the draft guidance addresses step 3 (public disclosure), outlining indications, practical guidance, and considerations in an outline and/or summary format. Real-world examples demonstrating application of the framework are being developed as supplementary resources to the framework. Current work group activities include engaging bioethicists, media reporters and patient advocates to review and comment on the guidance to ensure that it is clear, consistent and practical. Discussion: The draft guidance provides a framework for standardized actions for HAI outbreak notification and disclosure in a stepwise fashion, modeling public health practices and grounded in bioethical principles. The final product will provide valuable, practical advice for effectively sharing information with affected or potentially affected individuals and their caregivers in a timely manner.
Funding: None
Disclosures: Dawn Terashita reports that her spouse has received honoraria rom the speaker’s bureaus of Novo Nordisk and Abbott.