BoosTexter: A Boosting-based Systemfor Text Categorization
Machine Learning - Special issue on information retrieval
Knowledge Discovery in Multi-label Phenotype Data
PKDD '01 Proceedings of the 5th European Conference on Principles of Data Mining and Knowledge Discovery
Google's PageRank and Beyond: The Science of Search Engine Rankings
Google's PageRank and Beyond: The Science of Search Engine Rankings
Multilabel Neural Networks with Applications to Functional Genomics and Text Categorization
IEEE Transactions on Knowledge and Data Engineering
ML-KNN: A lazy learning approach to multi-label learning
Pattern Recognition
Random k-Labelsets: An Ensemble Method for Multilabel Classification
ECML '07 Proceedings of the 18th European conference on Machine Learning
The Top Ten Algorithms in Data Mining
The Top Ten Algorithms in Data Mining
Classifier Chains for Multi-label Classification
ECML PKDD '09 Proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases: Part II
Multi-label learning by exploiting label dependency
Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining
Mr.KNN: soft relevance for multi-label classification
CIKM '10 Proceedings of the 19th ACM international conference on Information and knowledge management
WEKA---Experiences with a Java Open-Source Project
The Journal of Machine Learning Research
RW.KNN: a proposed random walk KNN algorithm for multi-label classification
Proceedings of the 4th workshop on Workshop for Ph.D. students in information & knowledge management
| Hi-index | 0.00 |
Single-label classification refers to the task to predict an instance to be one unique label in a set of labels. Different from single-label classification, for multi-label classification, one instance is associated with one or more labels in a set of labels simultaneously. Various works have focused on the algorithms for those two types of classification. Since the ranking problem is always coexisting with the classification problem, and traditional researches mainly assume the uniform distribution for the instances, in this paper, we propose a new perspective for the ranking problem. With the assumption that the distribution for the instance is not uniform, different instances have different influences for the distribution, the Instance-Ranking algorithm is presented. With the Instance- Ranking algorithm, the famous K-nearest-neighbors (KNN) algorithm is modified to confirm the validity of our algorithm. Lastly, the Instance-Ranking algorithm is combined with the ML.KNN algorithm for multi-label classification. Experiment with different datasets show that our Instance-Ranking algorithm achieves better performance than the original state-of-art algorithm such as KNN and ML.KNN.