Similarity to Molecules in the Training Set Is a Good Discriminator for Prediction Accuracy in QSAR.
Robert P. SheridanBradley P. FeustonVladimir N. MaiorovSimon K. KearsleyPublished in: J. Chem. Inf. Model. (2004)
Keyphrases
- prediction accuracy
- training set
- predictive accuracy
- improve the prediction accuracy
- similarity measure
- test set
- predictive power
- web page prediction
- ensemble methods
- supervised learning
- data sets
- classification accuracy
- test data
- nearest neighbor
- distance measure
- active learning
- cross validation
- feature space
- face images
- improved accuracy
- training data
- semantic similarity
- recommender systems
- estimation accuracy
- drug discovery
- classification algorithm
- user ratings
- information retrieval
- training samples