Mining opinion summarizations using convolutional neural networks in Chinese microblogging systems

With recent advances in health systems, the amount of health data is expanding rapidly in various formats. This data originates from many new sources including digital records, mobile devices, and wearable health devices. Big health data offers more opportunities for health data analysis and enhancement of health services via innovative approaches. The objective of this research is to develop a framework to enhance health prediction with the revised fusion node and deep learning paradigms. Fusion node is an information fusion model for constructing prediction systems. Deep learning involves the complex application of machine-learning algorithms, such as Bayesian fusions and neural network, for data extraction and logical inference. Deep learning, combined with information fusion paradigms, can be utilized to provide more comprehensive and reliable predictions from big health data. Based on the proposed framework, an experimental system is developed as an illustration for the framework implementation.

show abstract

“…. , } ( is the number of features), is the number of filters, and denotes the weight matrix of the fully connected layer [47].…”

Section: Aligned Data Nmentioning

confidence: 99%

Enhancing Health Risk Prediction with Deep Learning on Big Data and Revised Fusion Node Paradigm

Zhong

Xiao

2017

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…Opinions are aspect-sentiment pairs that summarize a user’s sentiment toward a product at a fine granularity. Opinion summarization modeling aims to automatically mine aspect words and their corresponding sentiment words [ 28 ]. The model consists of the following two steps.…”

Section: Methodsmentioning

confidence: 99%

Understanding Users’ Vaping Experiences from Social Media: Initial Study Using Sentiment Opinion Summarization Techniques

Li¹,

Wang²,

Liu³

et al. 2018

J Med Internet Res

Self Cite

View full text Add to dashboard Cite

BackgroundE-liquid is one of the main components in electronic nicotine delivery systems (ENDS). ENDS review comments could serve as an early warning on use patterns and even function to serve as an indicator of problems or adverse events pertaining to the use of specific e-liquids—much like types of responses tracked by the Food and Drug Administration (FDA) regarding medications.ObjectiveThis study aimed to understand users’ “vaping” experience using sentiment opinion summarization techniques, which can help characterize how consumers think about specific e-liquids and their characteristics (eg, flavor, throat hit, and vapor production).MethodsWe collected e-liquid reviews on JuiceDB from June 27, 2013 to December 31, 2017 using its public application programming interface. The dataset contains 27,070 reviews for 8058 e-liquid products. Each review is accompanied by an overall rating and a set of 4 aspect ratings of an e-liquid, each on a scale of 1-5: flavor accuracy, throat hit, value, and cloud production. An iterative dichotomiser 3 (ID3)-based influential aspect analysis model was adopted to learn the key elements that impact e-liquid use. Then, fine-grained sentiment analysis was employed to mine opinions on various aspects of vaping experience related to e-liquids.ResultsWe found that flavor accuracy and value were the two most important aspects that affected users’ sentiments toward e-liquids. Of reviews in JuiceDB, 67.83% (18,362/27,070) were positive, while 12.67% (3430/27,070) were negative. This indicates that users generally hold positive attitudes toward e-liquids. Among the 9 flavors, fruity and sweet were the two most popular. Great and sweet tastes, reasonable value, and strong throat hit made users satisfied with fruity and sweet flavors, whereas “strange” tastes made users dislike those flavors. Meanwhile, users complained about some e-liquids’ steep or expensive prices, bad quality, and harsh throat hit. There were 2342 fruity e-liquids and 2049 sweet e-liquids. There were 55.81% (1307/2342) and 59.83% (1226/2049) positive sentiments and 13.62% (319/2342) and 12.88% (264/2049) negative sentiments toward fruity e-liquids and sweet e-liquids, respectively. Great flavors and good vapors contributed to positive reviews of fruity and sweet products. However, bad tastes such as “sour” or “bitter” resulted in negative reviews. These findings can help businesses and policy makers to further improve product quality and formulate effective policy.ConclusionsThis study provides an effective mechanism for analyzing users’ ENDS vaping experience based on sentiment opinion summarization techniques. Sentiment opinions on aspect and products can be found using our method, which is of great importance to monitor e-liquid products and improve work efficiency.

show abstract

“…Afterwards, the aggregator generates a report containing each aspect and its counts of positive and negative sentiment sentences. Likewise, Li et al [174] tackle opinion summarization in Chinese microblogs using a CNN and TextRank (TR) + maximal marginal relevance (MMR). CNN automatically learns the representative features from the input text.…”

Section: Applicationmentioning

confidence: 99%

Sentiment analysis using deep learning approaches: an overview

Habimana

et al. 2019

Sci. China Inf. Sci.

146

View full text Add to dashboard Cite

show abstract

Mining opinion summarizations using convolutional neural networks in Chinese microblogging systems

Cited by 64 publications

References 27 publications

Enhancing Health Risk Prediction with Deep Learning on Big Data and Revised Fusion Node Paradigm

Enhancing Health Risk Prediction with Deep Learning on Big Data and Revised Fusion Node Paradigm

Understanding Users’ Vaping Experiences from Social Media: Initial Study Using Sentiment Opinion Summarization Techniques

Sentiment analysis using deep learning approaches: an overview

Contact Info

Product

Resources

About