Performance Analysis of Different Neural Networks for Sentiment Analysis on IMDb Movie Reviews

Haque, Masudul; Lima, Salma Akter; Mishu, Sadia Zaman

doi:10.1109/icecte48615.2019.9303573

Cited by 29 publications

(15 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although a deep learning algorithm is merged with CNN to analyze sentiments more effectively. The same algorithms of LSTM and CNN are compared in [22] to measure the performance and accuracy of analyzing sentiments retrieved from movie ranking evaluations. The dataset is formalized according to the mechanism of each algorithm.…”

Section: Supervised Sentiment Analysismentioning

confidence: 99%

“…English Pos -Neg Deep Learning[17] Using deep learning to analyze Twitter dataset.English 3-Weight Deep Learning[18] Polarity prediction based on supervised learning.Chinese 3-Weight Deep Learning[19] Applying deep learning methods for analyzing sentiments automatically.English Pos -Neg Deep Learning[20] Applying classification using deep learning and machine learning methods.English Pos -Neg Supervised + Deep Learning[21] Merging Bi-LSTM-CNN to analyze sentiments. English 3-Weight Deep Learning[22] Merging LSTM with CNN to analyze reviews.English Pos -Neg Deep Learning[23] Applying CNN algorithm for determining the category or term polarity of the sentence aspect.English 3-Weight Deep Learning[24] Building a sentiment lexicon for five-weight classification using Genetic algorithm.English 5-Weight Genetic Algorithm[25] Applying HMM on sentences for predicting the polarity of sentiments.English 3-Weight Hidden Markov Model[26] Assigning a polarity score for each term to determine the overall polarity of the sentence. Weight Supervised/ Unsupervised[27] Creating a lexicon using Hadoop for storing sentences.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Enhanced Sentiment Analysis Algorithms for Multi-Weight Polarity Selection on Twitter Dataset

Mostafa¹

2023

Intelligent Automation &Amp; Soft Computing

View full text Add to dashboard Cite

Sentiment analysis is based on the orientation of user attitudes and satisfaction towards services and subjects. Different methods and techniques have been introduced to analyze sentiments for obtaining high accuracy. The sentiment analysis accuracy depends mainly on supervised and unsupervised mechanisms. Supervised mechanisms are based on machine learning algorithms that achieve moderate or high accuracy but the manual annotation of data is considered a time-consuming process. In unsupervised mechanisms, a lexicon is constructed for storing polarity terms. The accuracy of analyzing data is considered moderate or low if the lexicon contains small terms. In addition, most research methodologies analyze datasets using only 3-weight polarity that can mainly affect the performance of the analysis process. Applying both methods for obtaining high accuracy and efficiency with low user intervention during the analysis process is considered a challenging process. This paper provides a comprehensive evaluation of polarity weights and mechanisms for recent sentiment analysis research. A semi-supervised framework is applied for processing data using both lexicon and machine learning algorithms. An interactive sentiment analysis algorithm is proposed for distributing multi-weight polarities on Arabic lexicons that contain high morphological and linguistic terms. An enhanced scaling algorithm is embedded in the multi-weight algorithm to assign recommended weight polarities automatically. The experimental results are conducted on two datasets to measure the overall accuracy of proposed algorithms that achieved high results when compared to machine learning algorithms.

show abstract

Section: Supervised Sentiment Analysismentioning

confidence: 99%

mentioning

confidence: 99%

Enhanced Sentiment Analysis Algorithms for Multi-Weight Polarity Selection on Twitter Dataset

Mostafa¹

2023

Intelligent Automation &Amp; Soft Computing

View full text Add to dashboard Cite

show abstract

“…Önerilen modelde %91 oranında bir F1 değeri elde edilmiştir. Yapılan bu çalışmada araştırmacılar ESA ağlarının LSTM ağlarından daha iyi sonuç ürettiğini belirtmişlerdir [8].…”

Section: Literatür İncelemesiunclassified

Performing Sentiment Analysis with the Proposed Artificial Intelligence-Based Method Using Movie Commentaries

Yıldırım

2022

Fırat Üniversitesi Mühendislik Bilimleri Dergisi

View full text Add to dashboard Cite

With the developing technology, social media, forum sites and blocks have been widely used. People are now using these channels very widely and they share their feelings and thoughts in these environments. Therefore, natural language processing applications have started to become a more popular topic with each passing day. One of the most popular topics in natural language processing is sentiment analysis. In sentiment analysis, subjective information is extracted by making examinations according to certain criteria. In this study, the IMDB data set was used to perform sentiment analysis. The IMDB dataset is one of the largest datasets on this subject, consisting of movie reviews. This dataset contains users' comments about movies. In the study, firstly, the data preprocessing step was carried out. Then, the prepared data set was classified in classical machine learning classifiers and the proposed ESA-based model. The proposed ESA-based model was more successful than the classical machine learning classifiers in analyzing the texts in the IMDB dataset, and the proposed deep model achieved an accuracy of 85.57%.

show abstract

“…Hence, much research studied sentiment analysis comparing different architectures to get to the one with highest performance. Although the results slightly differ in accuracy and F1-score results, CNN architecture seemed to be the one outperforming other architecture like LSTM and LSTM-CNN as shown by these studies [1][2][3]5,8,[13][14][15]19,24]. Tokenization in one of the main data pre-processing before training or testing the CNN model.…”

Section: Introductionmentioning

confidence: 99%

“…To test this hypothesis, rebuilding experiments with a typo-less text, an ordinary text classification, was the start to get a benchmark for the initial results [2]. Then gradually introduce and increase percentage of typo was the way to see how both tokenization levels performs.…”

Section: Introductionmentioning

confidence: 99%

Effect of typos on text classification accuracy in word and character tokenization

Peter E. Shawky,

Saleh Mesbah ElKaffas,

Shawkat K Guirguis

2024

ARASET

View full text Add to dashboard Cite

To train a machine to “sense” a users’ feelings through writings (sentiment analysis) has become a crucial process in several domains: marketing, research, surveys and more. Nevertheless in times of crisis like COVID. Typo is one of the underestimated challenges processing user-generated text (comments, tweets, ..etc), it affects both learning and evaluation processes. Word tokenization outcome changes drastically even with a single character change, hence as expected, experiments have shown significant accuracy decreases due to typo. Adding a spelling correction as preprocessing layer, building one for every language, is a very time and resources expensive solution, a huge challenge against large data and real-time processing. Alternatively, a CNN model consuming the same text, once tokenized on characters level and once on words level while inducing typo, showed that as the typo percentage approaches 10% of the text, the results with characters tokens surpasses words tokens. Finally, on %30 typo of the text, the model consuming characters tokenization outperformed itself with the word level by a significant %22.3 in accuracy and %24.9 in F1-Score, using the same exact model. This approach in solving the inevitable typo challenge in NLP proved to be of significant practicality, saving huge resources versus using a spelling-correction model beforehand. It also removes a blocker challenge in front of real-time processing of user-generated text while preserving acceptable accuracy results.

show abstract

Performance Analysis of Different Neural Networks for Sentiment Analysis on IMDb Movie Reviews

Cited by 29 publications

References 10 publications

Enhanced Sentiment Analysis Algorithms for Multi-Weight Polarity Selection on Twitter Dataset

Enhanced Sentiment Analysis Algorithms for Multi-Weight Polarity Selection on Twitter Dataset

Performing Sentiment Analysis with the Proposed Artificial Intelligence-Based Method Using Movie Commentaries

Effect of typos on text classification accuracy in word and character tokenization

Contact Info

Product

Resources

About