TITLE

Multilabel classification with meta-level features in a learning-to-rank framework

AUTHOR(S)
Yang, Yiming; Gopal, Siddharth
PUB. DATE
July 2012
SOURCE
Machine Learning;Jul2012, Vol. 88 Issue 1/2, p47
SOURCE TYPE
Academic Journal
DOC. TYPE
Article
ABSTRACT
Effective learning in multi-label classification (MLC) requires an appropriate level of abstraction for representing the relationship between each instance and multiple categories. Current MLC methods have focused on learning-to-map from instances to categories in a relatively low-level feature space, such as individual words. The fine-grained features in such a space may not be sufficiently expressive for learning to rank categories, which is essential in multi-label classification. This paper presents an alternative solution by transforming the conventional representation of instances and categories into meta-level features, and by leveraging successful learning-to-rank retrieval algorithms over this feature space. Controlled experiments on six benchmark datasets using eight evaluation metrics show strong evidence for the effectiveness of the proposed approach, which significantly outperformed other state-of-the-art methods such as Rank-SVM, ML-kNN (Multi-label kNN), IBLR-ML (Instance-based logistic regression for multi-label classification) on most of the datasets. Thorough analyses are also provided for separating the factors responsible for the improved performance.
ACCESSION #
76517364

 

Related Articles

  • Statistical topic models for multi-label document classification. Rubin, Timothy; Chambers, America; Smyth, Padhraic; Steyvers, Mark // Machine Learning;Jul2012, Vol. 88 Issue 1/2, p157 

    Machine learning approaches to multi-label document classification have to date largely relied on discriminative modeling techniques such as support vector machines. A drawback of these approaches is that performance rapidly drops off as the total number of labels and the number of labels per...

  • A Study on Document Classification using Machine Learning Techniques. Thaoroijam, Kabita // International Journal of Computer Science Issues (IJCSI);Mar2014, Vol. 11 Issue 2, p217 

    With the explosion of information fuelled by the growth of the World Wide Web it is no longer feasible for a human observer to understand all the data coming in or even classify it into categories. With this growth of information and simultaneous growth of available computing power automatic...

  • InfoSift: A Novel, Mining-Based Framework for Document Classification. CHAKRAVARTHY, SHARMA; AERY, MANU; VENKATACHALAM, ARAVIND; TELANG, ADITYA // International Journal of Next-Generation Computing;Jul2014, Vol. 5 Issue 2, p84 

    A number of approaches, including machine learning, probabilistic, and information retrieval, have been proposed for classifying/retrieving documents where mainly words from the documents are used without considering any potential structural properties of the document. These techniques do not...

  • Document Classification Using Support Vector Machine. Mayor, Shweta; Pant, Bhasker // International Journal of Engineering Science & Technology;Apr2012, Vol. 4 Issue 4, p1741 

    Information like NEWS FEEDS is generally stored in the form of documents and files created on the basis of daily occurrence in the world. Classifying an unstructured text in these large document corpora has become cumbersome. Efficiently and effectively retrieving and categorizing these document...

  • Imbalanced Data Classification Based on AdaBoost-SVM. Li Peng; Bi Ting-ting; Yu Xiao-yang; Li Si-ben // International Journal of Database Theory & Application;2014, Vol. 7 Issue 5, p85 

    The classification of imbalanced data is one of the most challenging problems in data mining and machine learning research. Imbalanced dataset is a form that exists in reality area, which describes truly and objectively the essential characters of something. There will appear paucity of data and...

  • A Novel Text Classification Algorithm Based on Matrix Projection Method. Jiang Zhong; Lin Su; Qigan Sun // Advances in Information Sciences & Service Sciences;Apr2013, Vol. 5 Issue 7, p427 

    In this paper we presented a novel text classification algorithm based on matrix projection method. Matrix projection is used to solve the feature selection problem, which could improve the efficiency and accuracy of classification. We defined a novel matrix operation to calculate weight of each...

  • CONTECT BASED AUTOMATIC CLASSIFICATION OF RESEARCH ARTICLES. Khan, Shahid; Faheem Khan, Muhammad; Khan, Aurangzeb; Ullah Khan, Aziz; Ali, Shaukat // Science International;2014, Vol. 26 Issue 5, p2495 

    Research Article Classification is mainly concern with Document Classification process. Content of the article is used as a "Bag of Word" BOW with term frequency. The vector notation represents the bag-of-words. Various supervised and un-supervised learning techniques are used for classification...

  • Synergy of multi-label hierarchical ensembles, data fusion, and cost-sensitive methods for gene functional inference. Cesa-Bianchi, Nicolò; Re, Matteo; Valentini, Giorgio // Machine Learning;Jul2012, Vol. 88 Issue 1/2, p209 

    Gene function prediction is a complex multilabel classification problem with several distinctive features: the hierarchical relationships between functional classes, the presence of multiple sources of biomolecular data, the unbalance between positive and negative examples for each class, the...

  • CiteSeerX: AI in a Digital Library Search Engine. Jian Wu; William, Kyle; Hung-Hsuan Chen; Khabsa, Madian; Caragea, Cornelia; Tuarob, Suppawong; Ororbia, Alexander; Jordan, Douglas; Mitra, Prasenjit; Lee Giles, C. // AI Magazine;Fall2015, Vol. 36 Issue 3, p35 

    CiteSeerX is a digital library search engine that provides access to more than 5 million scholarly documents with nearly a million users and millions of hits per day. We present key AI technologies used in the following components: document classification and deduplication, document and citation...

Share

Read the Article

Courtesy of NEW JERSEY STATE LIBRARY

Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics