Designing online user contribution environments: A framework for formal analysis
Abstract
Statement of Work: The problem of designing effective online user contribution environments is one of central importance to fully harnessing the power of the Internet. The goal of this research is to create a sound formal foundation for the analysis and design of online contribution systems---such as social computing systems, platforms for online labor and crowdwork, and online peer-to-peer exchange economies---whose outcomes hinge on individual users decisions. Objective: The value created by the Internet now increasingly comes from systems centered around user contribution, whether explicit, as in user-generated content and social computing platforms, online labor markets, and platforms for microtasks, or implicit contribution via data collected from users that are used to provide improved recommendations and information. The users in these systems whose decisions determine the system s success of failure are economic agents with their own costs and benefits to participation and contribution. Given that users will make the choices that best trade off their own costs and benefits, how can one design environments so as to steer these individual decisions towards `good outcomes for the system? The primary focus of this research is on using formal analysis to inform the design of online environments that rely on user participation and contribution. Approach: The first component of such research is to formulate questions regarding design, namely how interfaces and (implicit or explicit) reward structures should be designed to steer user decisions towards choices that optimize system objectives, in a formal framework, and create an abstract model that captures these aspects, and at the same time also is tractable for mathematical analysis. Once such a model is in place, addressing the design problem will utilize techniques drawn primarily from economic theory and mechanism design, in particular relating to equilibrium analysis (such as of Nash equilibria or Bayes-Nash equilibria). Once the equilibria for a given mechanism have been determined, the question then turns to comparing amongst mechanisms in terms of which have the most desirable equilibrium outcomes. While the specific techniques used to achieve this vary from problem to problem, they broadly involve problem-solving utilizing an array of techniques from continuous and discrete mathematics. Overall Merit and ONR Mission/Relevance: The problem of designing mechanisms and environments whose outcomes rely on decison-making by agents is one of broad naval relevance. First, the general research themes that run through this proposal, namely behavioral design---laying the formal foundations for designing environments accounting for behavioral biases in individual decision making---and information as a design choice---understanding how the choice of information provided to agents affect their choices---can have impact across wide areas of decision-making, an area of general relevance to the Navy. Second, the specific application domains addressed by this proposal are of potentially high naval relevance as well: the field of human computation is exploding, with the creation of new platforms and systems to bring together crowds to enable innovations with increasing reach and power. The pervasive trend towards utilizing crowdsourcing and online platforms means that developing the formal foundations for the design of such online crowdbased environments---which rely on participation and contribution decisions by the individual users that make up the crowds---is one of central relevance to the success of these systems. Foundational research informing the design of such systems---for example, crowdsourcing contests for sourcing innovations of naval benefit, or the use of crowdwork for tasks such as image labeling or object detection---is of potential benefit, whether immediately or in the future, to the Navy.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 08, 2016
- Source ID
- N000141512335
Entities
People
- Arpita Ghosh
Organizations
- Cornell University
- Office of Naval Research
- United States Navy