Table 9 Summary of hybrid approach (combination only one of machine learning method with lexicon-based approach)

[107]

To perform sentiment analysis in customer review real word data

K-Mean Clustering + MPQA

Amazon review texts

Subjective expressions, positive, negative, neutral

[108]

To determine sentimental state of a person or a group of people using data mining

NB + lexicon-based analyser R platform

Twitter tweets

Emotions (anger, fear, disgust, surprise, happiness, and sadness), polarity (positive, negative, neutral)

[109]

To address the problem of estimating public opinion in social media content by proposing an aspect-based opinion mining model

NB + Wordnet

Online camera reviews

Positive, negative, neutral

[105]

To determined polarity of opinions toward a target word

To analyse and classify opinions

Neural-fuzzy network + SentiStrength data

Twitter

Positive polarity and negative polarity

. [110]

To build a customisable platform that collects the stream of relevant tweets generated by users, store them and do the sentiment analysis

SVM + SWN

Twitter, Heathrow and aircraft noise

Positive, negative or neutral

[111]

To classify tweets into three classes (positive, negative, neutral) using hybrid approach based on particular domain

Fuzzy logic + SentiWordNet

Tweets according or linked to a product, a hashtag or a movie review

Positive, negative or neutral

[112]

To find the scores of opinions from people’s reviews and derive conclusions

SVM + Wordnet

A movie review dataset has been collected from Twitter reviews

Negative and positive

[104]

To construct tourism emotion model

NB + sentiment dictionary constructed by Chen Bing

Microblog travel text online commentary

Positive, negative

[113]

To conduct emotion analysis in e-learning materials

SVM + SentiWordNet

E learners’ comments

Positive, negative, or neutral

[114]

To focus on sentiment analysis in financial newswire text

To classify sentiment expressed about certain companies in financial news articles

SVM + Dutch sentiment lexicons and Pattern lexicon

Internet Movie (IMDB) dataset

Positive and negative

[115]

To highlight the emotions and polarity communicated by an article liable to increase the prediction regarding its acceptability by the audience

RF + NRC suite of lexica: EmoLex11

Medium (the articles on the online publishing platform)

Negative and positive, joy, sadness, anger, fear, trust, surprise, disgust and anticipation

[116]

To monitor transportation activities (accidents, vehicles, street conditions, traffic volume, etc.)

To make a city-feature polarity map for travellers

Fuzzy ontology + SentiWordNet

Reviews from Twitter, Facebook and news

Positive, neutral or negative

[117]

To classify polarity of patient experiences of drugs using domain knowledge

Hybrid approach: FactNet, the knowledge base of polar facts

Drug reviews

Positive and negative

[118]

To use sentiment analysis and present a way to find relationships between tweets based on polarity and subjectivity

K-means algorithm + AFINN lexicon + TextBloB

Twitter data

Positive and negative

[119]

To propose a novel text representation model named Word2PLTS for short text sentiment analysis by introducing probabilistic linguistic terms sets (PLTSs) and relevant theory

SVM + SentiWordNet

Movie reviews (MR): Stanford Twitter Sentiment (STS): Tripadvisor reviews (TR)

Positive or negative

[120]

To compute the sentiments of social media posts

Fuzzy rule-based system + AFINN + VADER + SentiWordNet

Twitter datasets

Positive, negative or neutral

[121]

To extract user’s opinions and test them in two different datasets in English and Persian by introducing a part-of-speech graphical model

SVM + SentiWordNet,

Twitter, Iranian stock market

Positive or negative

[122]

To study Polarity Aggregation Model performance by extracting aspects of monument reviews and assigning to them the aggregated polarities

Deep Learning SAMs

Tripadvisor, English reviews

Positive, negative or neutral

[123]

To address the new methodology for dynamic modelling of customer preferences based on online customer reviews

Fuzzy + SentiWordNet

The online customer reviews of competitive hair dryers (Amazon.com)

Positive, neutral, and negative

[124]

To focus sentimental analysis on "times of India" movie review database

RF + SentiWordNet

Movie review dataset

Positive, negative and neutral