Predicting Mental Health
Predicting Mental Health from Twitter Accounts
This cross-disciplinary study between CS and Psychology researchers examines the extent to which the accounts a user follows on Twitter can be used to predict individual differences in self-reported anxiety, depression, post-traumatic stress, and anger.
The past decade has seen rapid growth in research linking stable psychological characteristics (i.e., traits) to digital records of online behavior in Online Social Networks (OSNs) like Facebook and Twitter, which has implications for basic and applied behavioral sciences. Findings indicate that a broad range of psychological characteristics can be predicted from various behavioral residue online, including language used in posts on Facebook (Park et al., 2015) and Twitter (Reece et al., 2017), and which pages a person ‘likes’ on Facebook (e.g., Kosinski, Stillwell, & Graepel, 2013). The present study examined the extent to which the accounts a user follows on Twitter can be used to predict individual differences in self-reported anxiety, depression, post-traumatic stress, and anger. Followed accounts on Twitter offer distinct theoretical and practical advantages for researchers; they are potentially less subject to overt impression management and may better capture passive users. Using an approach designed to minimize overfitting and provide unbiased estimates of predictive accuracy, our results indicate that each of the four constructs can be predicted with modest accuracy (out-of-sample R’s of approximately .2). Exploratory analyses revealed that anger, but not the other constructs, was distinctly reflected in followed accounts, and there was some indication of bias in predictions for women (vs. men) but not for racial/ethnic minorities (vs. majorities). We discuss our results in light of theories linking psychological traits to behavior online, applications seeking to infer psychological characteristics from records of online behavior, and ethical issues such as algorithmic bias and users’ privacy.
- Cory K. Costello (Psychology)
- Sanjay Srivastava (Psychology)
- Saed Rezayi (UO)
- Prof. Reza Rejaie (UO)
This material is based upon work supported by the National Science Foundation (NSF) Award 1551817. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF.
Understanding Incentivised Amazon Reviews
This study examines the problem of detecting and characterizing incentivized reviews in two primary categories of Amazon products.
During the past few years, sellers have increasingly offered discounted or free products to selected reviewers of e- commerce platforms in exchange for their reviews. Such incentivized (and often very positive) reviews can improve the rating of a product which in turn sways other users’ opinions about the product. Despite their importance, the prevalence, characteristics, and the influence of incentivized reviews in a major e-commerce platform have not been systematically and quantitatively studied.
This study examines the problem of detecting and characterizing incentivized reviews in two primary categories of Amazon products. We describe a new method to identify Explicitly Incentivized Reviews (EIRs) and then collect a few datasets to capture an extensive collection of EIRs along with their associated products and reviewers. We show that the key features of EIRs and normal reviews exhibit different characteristics. Furthermore, we illustrate how the prevalence of EIRs has evolved and been affected by Amazon’s ban. Our examination of the temporal pattern of submitted reviews for sample products reveals promotional campaigns by the corresponding sellers and their effectiveness in attracting other users. Finally, we demonstrate that a classifier that is trained by EIRs (without explicit keywords) and normal reviews can accurately detect other EIRs as well as implicitly incentivized reviews. Overall, this analysis sheds an insightful light on the impact of EIRs on Amazon products and users.
- Characterizing the Dynamics and Evolution of Incentivized Online Reviews on Amazon
Soheil Jamshidi, Reza Rejaie, Jun Li
Springer Social Network Analysis and Mining, 9(1) May 2019. DOI: 10.1007/s13278-019-0563-0
- Soheil Jamshidi (UO)
- Prof. Reza Rejaie (UO)
- Prof. Jun Li (UO)
Coarse View of Graphs
Inferring Multi-Resolution Views of Very Large Graphs
This study presents a new method to determine the coarse view of a large graph.
This study presents a simple framework, called WalkAbout, to infer a coarse view of connectivity in very large graphs; that is, identify well-connected “regions” with different edge densities and determine the corresponding inter- and intra- region connectivity. We leverage the transient behavior of many short random walks (RW) on a large graph that is assumed to have regions of varying edge density but whose structure is otherwise unknown. The key idea is that as RWs approach the mixing time of a region, the ratio of the number of visits by all RWs to the degree for nodes in that region converges to a value proportional to the average node degree in that region. Leveraging this indirect sign of con- nectivity enables our proposed framework to effectively scale with graph size.
After describing the design of WalkAbout, we demonstrate the capabilities of WalkAbout by applying it to three major OSNs (i.e., Flickr, Twitter, and Google+) and obtaining a coarse view of their connectivity structure. In addition, we illustrate how the communities that are obtained by run- ning a popular community detection method on these OSNs stack up against the WalkAbout-discovered regions. Finally, we examine the “meaning” of the regions obtained by Walk- About, and demonstrate that users in the identified regions exhibit common social attributes.
- Inferring Coarse Views of Connectivity in Very Large Graphs
Reza Motamedi, Reza Rejaie, Daniel Lowd, Walter Willinger, Roberto Gonzalez
ACM Conference on Online Social Networks (COSN), Dublin, Ireland, October 2014 [acceptance rate 16%]
- Reza Motamedi (UO)
- Prof. Reza Rejaie (UO)
- Prof. Daniel Lowd (UO)
- Dr. Walter Willinger (NIKSUN Inc.)
- Roberto Gonzalez (UC3M)
This project is funded by the National Science Foundation (NSF) grant no. IIS-0917381 and IIS-1342477. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.