Skip to main content

Table 10 Summary of hybrid approach (combination more than one of machine learning method with lexicon-based approach)

From: Opinion mining for national security: techniques, domain applications, challenges and research opportunities

Reference

Objective

Method used in Hybrid Approach for Opinion mining

Materials

Output

[106]

To evaluate, analyse and classify the opinions on behalf of user tweets toward smart devices

NB + SVM + lexicon dictionary

Twitter tweets

Polarity: positive or negative and emotion: anger, joy, sadness, disgust, fear and surprise

[125]

To store, query and analyse streaming data

knowledge-based + machine-learning + 3-way classification process + SentiWordNet

Twitter dataset

Positive, negative and neutral

[126]

To examine the sentiment expression

To classify the polarity of the movie review on a scale and perform feature extraction and ranking

To train multi-label classifier to classify the movie review into its correct label

RF + DT + NB + k-NN + SentiWordNet

Rotten Tomatoes movie review dataset

Positive and negative

[127]

To provide an automatic and accurate polarity classification of Twitter messages

NB + SVM + DT (J48) + KNN + SentiWordNet

Twitter messages

Positive or negative

[128]

To study public emotions and opinions concerning the opening of new IKEA stores

EN + LR + NB + SVM + NN + RF + English sentiment dictionary

Twitter texts, IKEA-related topics

Positive and negative

[129]

To perform effective sentimental analysis and opinion mining of web reviews using various rule-based machine learning algorithms

DT + NB + SentiWordNet

Text reviews

Strong-positive, positive, weak-positive, neutral, weak-negative, negative and strong-negative

[130]

To shortlist words that help in sentiment cognition

Fuzzy entropy + k-means clustering, LSTM + SentiWordNet

Movie review datasets (IMDB)

Positive or negative

[103]

To employ an emotion detection technique for sentiment classification

NB + SVM + NNs, LogN, RF, CART + NRC emotion lexicon

Twitter

Positive, negative and neutral

[131]

To deploy the phrase level sentiment analysis to classify online reviews into positive and negative polarities

fuzzy entropy + k-means clustering + SentiWordNet lexicon

Movie review, Pang-Lee and the IMDB dataset

Positive and negative

[132]

To present a sentiment polarity detection approach that detects sentiment polarity of Bengali tweets

Multinomial NB + SMO(SVM)) + SentiWordNet + Indian sentiment lexicon

Bengali Tweets dataset

Positive, negative and neutral