Search Results for author:
Found 5 papers, 5 papers with code
Can you Trust the Trend: Discovering Simpson's Paradoxes in Social Data
We present a statistical method to automatically identify Simpson's paradox in data by comparing statistical trends in the aggregate data to those in the disaggregated subgroups.
Computers and Society
Using Simpson's Paradox to Discover Interesting Patterns in Behavioral Data
We describe a data-driven discovery method that leverages Simpson's paradox to uncover interesting patterns in behavioral data.
MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing
Existing popular methods for semi-supervised learning with Graph Neural Networks (such as the Graph Convolutional Network) provably cannot learn a general class of neighborhood mixing relationships.
Friendship Paradox Biases Perceptions in Directed Networks
For example, in an online network of a social media platform, the number of people who mention a topic in their posts---i. e., its global popularity---can be dramatically different from how people see it in their social feeds---i. e., its perceived popularity---where the feeds aggregate their friends' posts.
Social and Information Networks Physics and Society
DoGR: Disaggregated Gaussian Regression for Reproducible Analysis of Heterogeneous Data
Quantitative analysis of large-scale data is often complicated by the presence of diverse subgroups, which reduce the accuracy of inferences they make on held-out data.
