Incorporating conditional random fields and active learning to improve sentiment identification

Zhang, Kunpeng; Xie, Yusheng; Yang, Yi; Sun, Aaron; Liu, Hengchang; Choudhary, Alok

doi:10.1016/j.neunet.2014.04.005

Cited by 30 publications

(17 citation statements)

References 33 publications

(37 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Features were extracted using newly proposed Entropy Weighted Genetic Algorithm (EWGA). The first experiment, conducted on movie review dataset [33,167] Along with these notable feature selection based studies, we surveyed others involving syntactic [92,130,171,174,178,190], semantic [32,92,48,60,69,109,155,169,172,190], stylistic [116,117,118,127], and information gain [75,165] based methods.…”

Section: Preprocessingmentioning

confidence: 99%

“…Non-ontological approaches can further be divided into topic modeling based [107,185,189,191,240] and non-topic modeling based approaches [25,37,59,67,186,190,193,218,241,243].…”

Section: Non-ontological Approachmentioning

confidence: 99%

“…The proposed model can further be utilized to discover a set of topics with shared features in a hierarchical structure. Zhang et al [190] suggested context-based sentiment classification method. They incorporated CRF with several syntactic and semantic features like similarity to neighboring sentences, word positions, comparative and superlative POS, conjunction terms, +ve/-ve emotions etc.…”

Section: Non-ontological Approachmentioning

confidence: 99%

See 2 more Smart Citations

A survey on opinion mining and sentiment analysis: Tasks, approaches and applications

Ravi

2015

Knowledge-Based Systems

1,074

528

View full text Add to dashboard Cite

Section: Preprocessingmentioning

confidence: 99%

“…Non-ontological approaches can further be divided into topic modeling based [107,185,189,191,240] and non-topic modeling based approaches [25,37,59,67,186,190,193,218,241,243].…”

Section: Non-ontological Approachmentioning

confidence: 99%

Section: Non-ontological Approachmentioning

confidence: 99%

See 1 more Smart Citation

A survey on opinion mining and sentiment analysis: Tasks, approaches and applications

Ravi

2015

Knowledge-Based Systems

1,074

528

View full text Add to dashboard Cite

“…We want to independently sample each bin from the histogram whose accuracy is to be evaluated. Within each histogram bin, we select users for verification like how active learning sample selection [14]. The upcoming selections depend on the description string of the previously selected users in that bin and whether our system tags them correctly into this bin.…”

Section: B Tagging Recall and Accuracymentioning

confidence: 99%

Estimating Online User Location Distribution without GPS Location

Xie

Agrawal

Choudhary

2014

2014 IEEE International Conference on Data Mining Workshop

Self Cite

View full text Add to dashboard Cite

We focus on the problem of offline user location estimation using online information, particularly for the application of TV segment advertising. Unlike previous works, the proposed method does not assume GPS information, but works with loosely structured information such as English location description. We propose to use a neural language model to capture the semantic similarity among the location descriptions. The language model can help reduce the otherwise expensive geolocating service lookups by internally resolving similar areas, neighborhoods, etc. onto the same description. We also propose a metric for comparing geodemographic histograms. This metric considers the demographic gap between the online world and the offline world. In the experiments section, we demonstrate the recall and accuracy of our language-based, GPS-free user location distribution estimation. In addition, we illustrate the effectiveness of the proposed distribution estimation metric.

show abstract

“…Since then CRFs have been successfully applied in a broad range of applied domains such as bioinformatics [5][6][7], computer vision [6][7][8][9][10][11][12][13][14][15][16], assessment of climate change impact [17], active learning [18,19], and natural language processing [4,[20][21][22]. A good introduction on CRFs is given in [23].…”

Section: Introductionmentioning

confidence: 99%

Conditional Random Fields for Pattern Recognition Applied to Structured Data

Burr

Skurikhin

2015

Algorithms

View full text Add to dashboard Cite

Pattern recognition uses measurements from an input domain, X, to predict their labels from an output domain, Y. Image analysis is one setting where one might want to infer whether a pixel patch contains an object that is "manmade" (such as a building) or "natural" (such as a tree). Suppose the label for a pixel patch is "manmade"; if the label for a nearby pixel patch is then more likely to be "manmade" there is structure in the output domain that can be exploited to improve pattern recognition performance. Modeling P(X) is difficult because features between parts of the model are often correlated. Therefore, conditional random fields (CRFs) model structured data using the conditional distribution P(Y|X = x), without specifying a model for P(X), and are well suited for applications with dependent features. This paper has two parts. First, we overview CRFs and their application to pattern recognition in structured problems. Our primary examples are image analysis applications in which there is dependence among samples (pixel patches) in the output domain. Second, we identify research topics and present numerical examples.

show abstract

Incorporating conditional random fields and active learning to improve sentiment identification

Cited by 30 publications

References 33 publications

A survey on opinion mining and sentiment analysis: Tasks, approaches and applications

A survey on opinion mining and sentiment analysis: Tasks, approaches and applications

Estimating Online User Location Distribution without GPS Location

Conditional Random Fields for Pattern Recognition Applied to Structured Data

Contact Info

Product

Resources

About