Sentiment Analysis on Chinese Hotel Reviews with Doc2Vec and Classifiers

Shuai, Qianjun; Huang, Yamei; Jin, Libiao; Pang, Lei

doi:10.1109/iaeac.2018.8577581

Cited by 24 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the classification task using machine learning, the main important thing is the feature selection [18]. This research using 10.000 data of Google Apps from Kaggle.…”

Section: Methodsmentioning

confidence: 99%

“…The method was designed according to the characteristics of the text. To represent the rank among the best approach in retrieving documents and labeling document, TF-IDF was used, Word2vec to obtain more accurate word vector, Doc2vec is one of the easiest ways is using an average of all words in the document to represent the feature of this document [18]. SVM, NB, and DT classifier were used in this paper for text classifier.…”

Section: Related Workmentioning

confidence: 99%

“…Fig. 3 shows two architectures from [18], the first is DMPV, and it will add a paragraph id to be trained with word vectors. This paragraph id contains information that is missing from the current word.…”

Section: ) Word2vecmentioning

confidence: 99%

“…Fig. 4 shows ways to input into the DBOW model is a paragraph id, predicting randomly sampled words in this document [18].…”

Section: ) Word2vecmentioning

confidence: 99%

“…It was implemented using the Sklearn library. 1) Support vector machine: Support Vector Machine (SVM) as a classifier is fully determined by a relatively small subset of the training instances, it is suitable for discrete data [18]. LinearSVC was used in this research as it is similar to SVM but with a kernel "linear" parameter and implemented in terms of liblinear rather than libsvm.…”

Section: Classificationmentioning

confidence: 99%

See 4 more Smart Citations

Optimizing the Hyperparameter of Feature Extraction and Machine Learning Classification Algorithms

Isa¹,

Suwandi²,

Pricilia³

2019

ijacsa

View full text Add to dashboard Cite

The process of assigning a quantitative value to a piece of text expressing a mood or effect is called Sentiment analysis. Comparison of several machine learning, feature extraction approaches, and parameter optimization was done to achieve the best accuracy. This paper proposes an approach to extracting comparison value of sentiment review using three features extraction: Word2vec, Doc2vec, Terms Frequency-Inverse Document Frequency (TF-IDF) with machine learning classification algorithms, such as Support Vector Machine (SVM), Naive Bayes and Decision Tree. Grid search algorithm is used to optimize the feature extraction and classifier parameter. The performance of these classification algorithms is evaluated based on accuracy. The approach that is used in this research succeeded to increase the classification accuracy for all feature extractions and classifiers using grid search hyperparameter optimization on varied pre-processed data.

show abstract

“…In the classification task using machine learning, the main important thing is the feature selection [18]. This research using 10.000 data of Google Apps from Kaggle.…”

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: ) Word2vecmentioning

confidence: 99%

“…Fig. 4 shows ways to input into the DBOW model is a paragraph id, predicting randomly sampled words in this document [18].…”

Section: ) Word2vecmentioning

confidence: 99%

Section: Classificationmentioning

confidence: 99%

See 3 more Smart Citations