Spectral Clustering-Based Semi-supervised Sentiment Classification

Li, Suke; Hao, Jinmei

doi:10.1007/978-3-642-35527-1_23

Cited by 6 publications

(8 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These negative indicators, just as in the previous work [17] [18], including "not", "no", "donot", "do not", "didn't", "did not", "was not", "wasn't", "isn't", "isn't", "weren't", "weren't", "doesn't", "doesn't", "hardly", "never", "neither", and "nor". Use training data set T and SVM to get the first sentiment classifier f 1 ;…”

Section: The Proposed Schemementioning

confidence: 53%

“…The Self-learning SVM is a bootstrap approach to learning as Algorithm 2 shows. The method is also used in the previous work [17] to do some comparisons. The algorithm iteratively selects the most likely correctly classified instances which are determined by their distances to classification hyperplane, and put them into the training data set.…”

Section: Compared Methodsmentioning

confidence: 99%

“…Each data category includes 1000 positive reviews and 1000 negative reviews, as Table 1 shows. These data sets also were studied in the previous publications [17] [18]. Reviews were also segmented into sentences.…”

Section: Experiments 41 Experiments Datamentioning

confidence: 99%

“…The previous work [17] also gave a spectral clustering-based semi-supervised sentiment classification method. In the work [18], the authors proposed a linear regression with rules-based approach to ranking product features according to their importance.…”

Section: Prelated Workmentioning

confidence: 99%

See 3 more Smart Citations

Sentiment Classification using Subjective and Objective Views

Li¹

2013

IJCA

Self Cite

View full text Add to dashboard Cite

This work proposes a new semi-supervised sentiment classification method by exploiting a large number of unlabeled instances to conduct sentiment classification for Web consumer reviews. In the proposed method every consumer review has two views: subjective view and objective view. The subjective view of a consumer review reflects the opinions expressed by opinion words, while the objective view is constructed by the remaining text features. This work is trying to combine two kinds of views to carry out sentiment classification. The method is based on the co-training framework which needs three basic sentiment classifiers to iteratively get the final sentiment classifier. In the proposed method, the first sentiment classifier is constructed using the common unigram features coming from consumer reviews. The second sentiment classifier is trained on the subjective views constructed by opinion words extracted from consumer reviews. The remaining text features of these reviews are used for obtaining the objective views which can be trained for the third classifier. Experimental results show the proposed method is effective, and it has better performance than the Self-learning SVM method. General TermsAlgorithm

show abstract

Section: The Proposed Schemementioning

confidence: 53%

Section: Compared Methodsmentioning

confidence: 99%

Section: Experiments 41 Experiments Datamentioning

confidence: 99%

Section: Prelated Workmentioning

confidence: 99%

See 2 more Smart Citations

Sentiment Classification using Subjective and Objective Views

Li¹

2013

IJCA

Self Cite

View full text Add to dashboard Cite

show abstract

“…A semisupervised machine learning-based sentiment classification method only labels a small number of training instances and some unlabeled instances. The problem is defined as the work [11]. We firstly prepare for a small training data set T = {(x 1 , y 1 ), (x 2 , y 2 ), …, (x n , y n )}, where x i is the training instance.…”

Section: Introduction To Our Methodsmentioning

confidence: 99%