Prediction models for network-linked data
From MaRDI portal
Publication:2415451
Abstract: Prediction algorithms typically assume the training data are independent samples, but in many modern applications samples come from individuals connected by a network. For example, in adolescent health studies of risk-taking behaviors, information on the subjects' social network is often available and plays an important role through network cohesion, the empirically observed phenomenon of friends behaving similarly. Taking cohesion into account in prediction models should allow us to improve their performance. Here we propose a network-based penalty on individual node effects to encourage similarity between predictions for linked nodes, and show that incorporating it into prediction leads to improvement over traditional models both theoretically and empirically when network cohesion is present. The penalty can be used with many loss-based prediction methods, such as regression, generalized linear models, and Cox's proportional hazard model. Applications to predicting levels of recreational activity and marijuana usage among teenagers from the AddHealth study based on both demographic covariates and friendship networks are discussed in detail and show that our approach to taking friendships into account can significantly improve predictions of behavior while providing interpretable estimates of covariate effects.
Recommendations
Cited in
(18)- A note on a dynamic network model with homogeneous structure
- Multiclass analysis and prediction with network structured covariates
- Bayesian reconstruction of Cartesian product graph signals with general patterns of missing data
- Prediction models with graph kernel regularization for network data
- Depth-based classification for relational data with multiple attributes
- Subset selection in network-linked data
- Detecting clusters in multivariate response regression
- Logistic regression with network structure
- Bayesian spatial homogeneous regression
- High-dimensional Gaussian graphical models on network-linked data
- Network regression with predictive clustering trees
- Spatial Tweedie exponential dispersion models: an application to insurance rate-making
- Integrating graph structures into kernel regression models
- Joint modeling of playing time and purchase propensity in massively multiplayer online role-playing games using crossed random effects
- A Unifying Generative Model for Graph Learning Algorithms: Label Propagation, Graph Convolutions, and Combinations
- Network vector autoregression with individual effects
- A Bayesian approach for subgroup analysis
- Modeling urban crime occurrences via network regularized regression
This page was built for publication: Prediction models for network-linked data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2415451)
