Abstract
We analyze a Relational Neighbor (RN) classifier, a simple relational\npredictive model that predicts only based on class labels of related neighbors,\nusing no learning and no inherent attributes.We show that it performs surprisingly\nwell by comparing it to more complex models such as Probabilistic Relational\nModels and Relational Probability Trees on three data sets from published work.\nWe argue that a simple model such as this should be used as a baseline to assess\nthe performance of relational learners.
Keywords
Simple (philosophy)Classifier (UML)Computer scienceArtificial intelligencePattern recognition (psychology)EpistemologyPhilosophy
Affiliated Institutions
Related Publications
Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
Many real-world applications require the prediction of long sequence time-series, such as electricity consumption planning. Long sequence time-series forecasting (LSTF) demands ...
Earth Structure from Free Oscillations and Travel Times
An extensive set of reliable gross Earth data has been inverted to obtain a new estimate of the radial variations of seismic velocities and density in the Earth. The basic data ...
Nonâalcoholic steatohepatitis: Definitions and pathogenesis
The term 'non-alcoholic steatohepatitis' or NASH was first used by Ludwig et al. in 1980 to describe 'the pathological and clinical features of non-alcoholic disease of the live...
Publication Info
- Year
- 2003
- Type
- report
- Citations
- 244
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
244
OpenAlex
Cite This
Sofus A. Macskassy,
Foster Provost
(2003).
A Simple Relational Classifier.
.
https://doi.org/10.21236/ada452802
Identifiers
- DOI
- 10.21236/ada452802