Solving multi-label text categorization problem using support vector machine approach with membership function

Wang, Tai-Yue; Chiang, Huei-Min

doi:10.1016/j.neucom.2011.07.001

Cited by 26 publications

(14 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It suffers from high dimensionality of feature space and huge memory burden. [12] proposes a modified one-against-one SVM classifier for multi-label text categorization using the SVM's predictions and probability, which is computationally expensive.…”

Section: Problem Transformation Methodsmentioning

confidence: 99%

Multi-label learning with discriminative features for each label

Zhang

Fang

2015

Neurocomputing

View full text Add to dashboard Cite

Section: Problem Transformation Methodsmentioning

confidence: 99%

Multi-label learning with discriminative features for each label

Zhang

Fang

2015

Neurocomputing

View full text Add to dashboard Cite

“…Based on this representation, scaling the dimensions of the feature vector with their respective inverse document frequency (IDF, which is applied as the log inverse of ω i ) led to an improved performance. According to the study, [2] IDF can be calculated from the total number of training documents (n) and the document frequency of the particular word ω i as shown in (1):…”

Section: Review Of Approaches In Document Classificationmentioning

confidence: 99%

“…Based on the standard feature vector representation of the text data, it was argued in the study [2] that the support vector machines are more appropriate for this type of setting. Different classification methods such as Bayes, SVM, C4.5 and kNN were applied on the Reuters-21578 and Ohsumed corpus [2] among which SVM was found to have superior prediction with considerable performance gain.…”

Section: Review Of Approaches In Document Classificationmentioning

confidence: 99%

“…[2] In this project we perform multi-class classification, in which a file is predicted into one of the predefined categories. Supervised learning models are widely applicable and can offer the insight about how the explanatory variables are related to the categorical response variable.…”

Section: Approaches To Document Classifica-tion Using Machine Learningmentioning

confidence: 99%

“…Multi-class labeled document classification is relatively challenging compared to the single-class labeled document. [2] In addition to supervised and unsupervised learning there exist a special form of learning known as semi-supervised learning (SSL). [3] SSL falls in between supervised and unsupervised learning.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multiclass patent document classification

Anne

Mishra

Hoque

et al. 2017

AIR

View full text Add to dashboard Cite

This article addresses patent document classification problem into fifteen different categories or classes, where some classes overlap with each other for practical reasons. For the development of the classification model using machine learning techniques, useful features have been extracted from the given documents. The features are used to classify patent document as well as to generate useful tag-words. The overall objective of this work is to systematize NASA's patent management, by developing a set of automated tools that can assist NASA to manage and market its portfolio of intellectual properties (IP), and to enable easier discovery of relevant IP by users. We have identified an array of methods that can be applied such as k-Nearest Neighbors (kNN), two variations of the Support Vector Machine (SVM) algorithms, and two tree based classification algorithms: Random Forest and J48. The major research steps in this paper consist of filtering techniques for variable selection, information gain and feature correlation analysis, and training and testing potential models using effective classifiers. Further, the obstacles associated with the imbalanced data were mitigated by adding pseudo-synthetic data wherever appropriate, which resulted in a superior SVM classifier based model.

show abstract