Crowdsourcing Predictors of Behavioral Outcomes

Abstract

Generating models from large data sets -- and determining which subsets of data to mine -- is becoming increasingly automated. However choosing what data to collect in the first place requires human intuition or experience, usually supplied by a domain expert. This paper describes a new approach to machine science which demonstrates for the first time that non-domain experts can collectively formulate features, and provide values for those features such that they are predictive of some behavioral outcome of interest. This was accomplished by building a web platform in which human groups interact to both respond to questions likely to help predict a behavioral outcome and pose new questions to their peers. This results in a dynamically-growing online survey, but the result of this cooperative behavior also leads to models that can predict user's outcomes based on their responses to the user-generated survey questions. Here we describe two web-based experiments that instantiate this approach: the first site led to models that can predict users' monthly electric energy consumption; the other led to models that can predict users' body mass index. As exponential increases in content are often observed in successful online collaborative communities, the proposed methodology may, in the future, lead to similar exponential rises in discovery and insight into the causal factors of behavioral outcomes

Tracking Post-trauma Psychopathology Using Mobile Applications: a Usability Study

“…Such qualitative information may also lead to new insights into the acute post-trauma period and improve assessment and tools. For example, mixed-method studies have used similar responses with machine learning to derive highly accurate measures for other constructs [29]. …”

Eliciting public concerns about an emerging energy technology: The case of unconventional shale gas development in the United States

“…The snowball sampling process based on seeds mined from searches of Facebook sites and Internet pages is one possible replicable approach. Similar approaches to identifying concerns are gaining interest, such as the use of crowdsourcing [44,45] and on-line discussion boards [46]. Both of these also rely on snowball sampling based on reaching out to a core seed of knowledgeable parties.…”

Performance Analysis for Crowdsourcing Context Submission using Hierarchical Clustering Algorithm and Classification

“…Existing work has shown that the cluster formation of crowdsourcing website context need further enhancement [1], [2], [3], [4], [5], [6], [7]. In [1], behavioral model is implemented which leads into dynamically growing online data which is sometimes noisy, missing values etc.…”

“…In [1], behavioral model is implemented which leads into dynamically growing online data which is sometimes noisy, missing values etc. resulting into difficulty into submission of crowdsourcing website context.…”