Instance-based prediction of real-valued attributes
Computational Intelligence
Instance-Based Learning Algorithms
Machine Learning
A Nearest Hyperrectangle Learning Method
Machine Learning
Possibilistic instance-based learning
Artificial Intelligence - Special issue: Fuzzy set and possibility theory-based methods in artificial intelligence
Selective Sampling for Nearest Neighbor Classifiers
Machine Learning
An integrated approach to bioprocess recipe design
Integrated Computer-Aided Engineering
On the use of surrounding neighbors for synthetic over-sampling of the minority class
SMO'08 Proceedings of the 8th conference on Simulation, modelling and optimization
Cho-k-NN: a method for combining interacting pieces of evidence in case-based learning
IJCAI'05 Proceedings of the 19th international joint conference on Artificial intelligence
Learning Instance-Specific Predictive Models
The Journal of Machine Learning Research
Instance selection with neural networks for regression problems
ICANN'12 Proceedings of the 22nd international conference on Artificial Neural Networks and Machine Learning - Volume Part II
| Hi-index | 0.00 |
Lazy learning methods for function prediction use different predictionfunctions. Given a set of stored instances, a similarity measure, and a novelinstance, a prediction function determines the value of the novel instance.A prediction function consists of three components: a positive integer kspecifying the number of instances to be selected, a method for selectingthe k instances, and a method for calculating the value of the novelinstance given the k selected instances. This paper introduces a novelmethod called k surrounding neighbor (k-SN) for intelligentlyselecting instances and describes a simple k-SN algorithm. Unlike k nearestneighbor (k-NN), k-SN selects k instancesthat surround the novel instance. We empirically compared k-SN withk-NN using the linearly weighted average and local weightedregression methods. The experimental results show that k-SNoutperforms k-NN with linearly weighted average and performs slightlybetter than k-NN with local weighted regression for the selecteddatasets.