FIGURE 2 - uploaded by Ali Feizollah
Content may be subject to copyright.
Structure of deep learning scheme.
Source publication
Twitter is a leading platform among social media networks. It allows microblogging of up to 140 characters for a single post. Owing to this characteristic, it is popular among users. People tweet about various topics, from daily life events to major incidents. Given the influence of this social media platform, the analysis of Twitter contents has b...
Context in source publication
Context 1
... general and basic structure of the deep learning scheme is shown in Figure 2. It consists of an input layer, which is the input data to the algorithms; hidden layers, in which the algorithm makes numerous mathematical calculations; and the output layer, which is the result of the calculations of the algorithms. ...
Similar publications
![FIGURE 1. A generic architecture of CNN based for text classification [31].](profile/Rachid-Ben-Said/publication/349939323/figure/fig1/AS:999624705839106@1615340510345/A-generic-architecture-of-CNN-based-for-text-classification-31_Q320.jpg)
![FIGURE 2. Architecture of a Bi-LSTM [34].](profile/Rachid-Ben-Said/publication/349939323/figure/fig2/AS:999624705847296@1615340510418/Architecture-of-a-Bi-LSTM-34_Q320.jpg)
Being one of the most widely used social media tools, Twitter is seen as an important source of information for acquiring people’s attitudes, emotions, views and feedbacks. Within this context, Twitter sentiment analysis techniques were developed to decide whether textual tweets express a positive or negative opinion. In contrast to lower classific...
![Fig 3: Heatmap of Yelp reviews with two extreme scores [29]](profile/Sifat-Ahmed/publication/343671575/figure/fig1/AS:924750666809345@1597489148628/Heatmap-of-Yelp-reviews-with-two-extreme-scores-29_Q320.jpg)
In today's world, social networks are the place where user share their views, emotions in their way. Social media, such as Twitter, Instagram, Facebook, etc. where millions of people express their views in their daily day-today life, which can be their sentiments and opinions or expressing emotions about a particular thing or their own. This gave r...
Social media is progressively exposed to issues of damaging conduct, for example, private assaults and cyberbullying. It is troublesome and time consuming to manually check which of the comments need to be blocked. Therefore, automating the process of identifying the abusive comments and blocking them would not only save time, but also provide safe...
Public concern detection provides potential guidance to the authorities for crisis management before or during a pandemic outbreak. Detecting people's concerns and attention from online social media platforms has been widely acknowledged as an effective approach to relieve public panic and prevent a social crisis. However, detecting concerns in tim...
This paper describes the research that our team, DA Master, did on the shared task HASOC, conducted by FIRE-2019, which involves identification of hate and offensive language in Twitter and Facebook posts. The task is divided into three sub-tasks. Our team conducted our experiments on an English dataset. We employed Machine Learning techniques like...
Citations
... Legal analytics, policymaking, and the creation of AI-powered legal tools all stand to benefit significantly from this breakthrough [35]. Feizollah, et al. [36] utilized CNN and LSTM algorithms to extract Twitter text and claimed 93.78% accuracy. ...
This research investigates the application of deep learning in sentiment analysis of Canadian maritime case law. It offers a framework for improving maritime law and legal analytic policy-making procedures. The automation of legal document extraction takes center stage, underscoring the vital role sentiment analysis plays at the document level. Therefore, this study introduces a novel strategy for sentiment analysis in Canadian maritime case law, combining sentiment case law approaches with state-of-the-art deep learning techniques. The overarching goal is to systematically unearth hidden biases within case law and investigate their impact on legal outcomes. Employing Convolutional Neural Network (CNN)- and long short-term memory (LSTM)-based models, this research achieves a remarkable accuracy of 98.05% for categorizing instances. In contrast, conventional machine learning techniques such as support vector machine (SVM) yield an accuracy rate of 52.57%, naïve Bayes at 57.44%, and logistic regression at 61.86%. The superior accuracy of the CNN and STM model combination underscores its usefulness in legal sentiment analysis, offering promising future applications in diverse fields like legal analytics and policy design. These findings mark a significant choice for AI-powered legal tools, presenting more sophisticated and sentiment-aware options for the legal profession.
... A method based on word sequencing [32] was employed, neglecting the consideration of individual word weights. This approach involved transforming the word sequence into a matrix, where the length represented the input size, and the height corresponded to the number of observations. ...
Sentiment analysis, a critical task in natural language processing, aims to automatically identify and classify the sentiment expressed in textual data. Aspect-level sentiment analysis focuses on determining sentiment at a more granular level, targeting specific aspects or features within a piece of text. In this paper, we explore various techniques for sentiment analysis, including traditional machine learning approaches and state-of-the-art deep learning models. Additionally, deep learning techniques has been utilized to identifying and extracting specific aspects from text, addressing aspect-level ambiguity, and capturing nuanced sentiments for each aspect. These datasets are valuable for conducting aspect-level sentiment analysis. In this article, we explore a language model based on pre-trained deep neural networks. This model can analyze sequences of text to classify sentiments as positive, negative, or neutral without explicit human labeling. To evaluate these models, data from Twitter's US airlines sentiment database was utilized. Experiments on this dataset reveal that the BERT, RoBERTA and DistilBERT model outperforms than the ML based model in accuracy and is more efficient in terms of training time. Notably, our findings showcase significant advancements over previous state-of-the-art methods that rely on supervised feature learning, bridging existing gaps in sentiment analysis methodologies. Our findings shed light on the advancements and challenges in sentiment analysis, offering insights for future research directions and practical applications in areas such as customer feedback analysis, social media monitoring, and opinion mining.
... CNNs excel in extracting spatial features, making them well suited for data like images with spatial patterns, while LSTMs specialize in capturing temporal dependencies, fitting perfectly for sequential data such as time series or text [39][40][41]. This coupling allows for hierarchical feature learning, where the CNNs extract features and the LSTMs sequentially process them for deeper insight [42,43]. Furthermore, this model exhibits acceptable computation times for predictions when compared to laboratory analysis. ...
In modern aquaculture, the focus is on optimizing production and minimizing environmental impact through the use of recirculating water systems, particularly in outdoor setups. In such systems, maintaining water quality is crucial for sustaining a healthy environment for aquatic life, and challenges arise from instrumentation limitations and delays in laboratory measurements that can impact aquatic animal production. This study aimed to predict key water quality parameters in an outdoor recirculation aquaculture system (RAS) for red tilapia aquaculture, including dissolved oxygen (DO), pH, total ammonia nitrogen (TAN), nitrite nitrogen (NO 2-N), and alkalinity (ALK). Initially, a random forest (RF) model was employed to identify significant factors for predicting each parameter, selecting the top three features from routinely measured parameters on the farm: DO, pH, water temperature (Temp), TAN, NO 2-N, and transparency (Trans). This approach aimed to streamline the analysis by reducing variables and computation time. The selected parameters were then used for prediction, comparing the performance of convolutional neural network (CNN), long short-term memory (LSTM), and CNN-LSTM models across different epochs (1000, 3000, and 5000). The results indicated that the CNN-LSTM model at 5000 epochs was effective in predicting DO, TAN, NO 2-N, and ALK, with high R 2 values (0.815, 0.826, 0.831, and 0.780, respectively). However, pH prediction showed lower efficiency with an R 2 value of 0.377.
... The output was very effective in classification and also in recognition. Another paper that has taken tweets about the tourist activities and cosmetics, that are allowed according to Islam, in the past few years on two different languages: English and Malay is done by A. Feizollah et al. [4] in 2019. They have used python language to analyse by finding in which language the tweets taken are written and they have used deep learning like RNN, CNN and LSTM for sentiment analysis. ...
... However, the limit is up to the previous seven days for data collection. A premium account, which charges a lot can provide data older than seven days [68]. Also, Twitter has features for filtration to get the data required. ...
Sentiment analysis is a part of natural language processing, along with text mining. Over the years, sentiment analysis has become a key study area for researchers and industries all over the world. The major goal of this review is to overview the papers that have found the sentiment of the data under study over the past few years. Also, the various techniques and methods that have enabled us to solve the problem of sentiment analysis are looked into and put forward briefly. The articles are looked into considering the area of sentiment analysis in which the contributions are made, be it sentiment detection, dataset creation, or transfer learning. This would help the researchers to curate advanced and accurate models and methods for analyzing sentiment. Additionally, they would be given a quick rundown of current developments in this area of research. With this in mind, the literature study is conducted taking into account various applications, various machine learning algorithms, the type of data utilized for analysis purposes, and various performance measurements. Challenges and gaps based on all the contributions studied are summarized.
... To address these limitations, deep learning, a cluster of multi-layer neural network algorithms have emerged as a promising sub-field of machine learning for Twitter sentiment analysis [32,33,34]. Several deep learning-based models, including Deep (Vanilla) Neural Networks (DNN) Ali et al. [32], Yasir et al. [34], Convolutional Neural Networks (CNN) [35,36,37,38], Recurrent Neural Networks (RNN) [39,40], and their variants such as Long Short-Term Memory (LSTM) [41,42,43,44], Gated Recurrent Units (GRU) and hybrid techniques have shown effectiveness in capturing the nuances of natural language and handling the noise and ambiguity present in Twitter data [35,36,37,38,39,40,41,42,43,44]. These models offer flexible solutions that enhance sentiment analysis performance by providing a better interpretation of the context and semantic meaning of text data. ...
... To address these limitations, deep learning, a cluster of multi-layer neural network algorithms have emerged as a promising sub-field of machine learning for Twitter sentiment analysis [32,33,34]. Several deep learning-based models, including Deep (Vanilla) Neural Networks (DNN) Ali et al. [32], Yasir et al. [34], Convolutional Neural Networks (CNN) [35,36,37,38], Recurrent Neural Networks (RNN) [39,40], and their variants such as Long Short-Term Memory (LSTM) [41,42,43,44], Gated Recurrent Units (GRU) and hybrid techniques have shown effectiveness in capturing the nuances of natural language and handling the noise and ambiguity present in Twitter data [35,36,37,38,39,40,41,42,43,44]. These models offer flexible solutions that enhance sentiment analysis performance by providing a better interpretation of the context and semantic meaning of text data. ...
... To gain better insights into recent years' advancements, the current survey bifurcates the DL literature into a taxonomy broadly categorized as Basic and Transformer-based. Basic DL models consist of Deep Neural Network (DNN) [32,33,34], Convolutional Neural Network (CNN), [35,36,37,38], Recurrent Neural Network (RNN) [40], Long Short-Term Memory (LSTM) [110] whereas Transformer-based includes BERT [111], RoBERTa [112], XLNet [113], and GPT [114] etc. Besides, these two major categories there are many DL-hybrid methods proposed by the research community for Twitter text sentiment analysis along with recent developments of Graphbased methods that are classified under the "other" category in the current study. The following sections detail the literature for each category of DL models. ...
Microblogging site Twitter (re-branded to X since July 2023) is one of the most influential online social media websites, which offers a platform for the masses to communicate, expresses their opinions, and shares information on a wide range of subjects and products, resulting in the creation of a large amount of unstructured data. This has attracted significant attention from researchers who seek to understand and analyze the sentiments contained within this massive user-generated text. The task of sentiment analysis (SA) entails extracting and identifying user opinions from the text, and various lexicon-and machine learning-based methods have been developed over the years to accomplish this. However, deep learning (DL)-based approaches have recently become dominant due to their superior performance. This study briefs on standard preprocessing techniques and various word embeddings for data preparation. It then delves into a taxonomy to provide a comprehensive summary of DL-based approaches. In addition, the work compiles popular benchmark datasets and highlights evaluation metrics employed for performance measures and the resources available in the public domain to aid SA tasks. Furthermore, the survey discusses domain-specific practical applications of SA tasks. Finally, the study concludes with various research challenges and outlines future outlooks for further investigation.
... Various studies have used Twitter conversation sources with different machine-learning approaches, especially in the halal domain. These include mining public opinion and gauging public sentiment towards halal tourism and halal cosmetics [9], halal food global [10], halal certification [11], [12], and halal terms [13]. Furthermore, the same study showed that the availability of information on social media regarding halal food sources played a significant role in Malaysians' choice to purchase halal products [14]. ...
... To reduce dimensionality and select the most informative features, feature selection techniques are employed in the study. These techniques help identify features that show a strong correlation with sentiment, thereby increasing the accuracy of sentiment classification (Feizollah et al., 2019). Feature selection methods such as mutual information and chi-square are commonly used in sentiment analysis. ...
As part of this study, a unique method for automatic real-time analysis of sentiment in product reviews for online shopping applications is provided. The main goal is to create a model that has a high level of accuracy and is able to categorize product reviews with either a positive, negative, or neutral sentiment. To achieve this goal, a combination of natural language processing (NLP) strategies and machine learning algorithms is employed. The first step in the process, known as pre-processing, cleans up the text data by removing any noise and then applies tokenization and stemming techniques to it so that significant features can be extracted. A variety of machine learning models, such as B. Support Vector Machines (SVM), Naive Bayes and Random Forest, are trained and evaluated using an extensive data set of labeled customer reviews of various products. The development of a web application was chosen as the implementation method for the system in order to enable easy integration with the online trading platform. This solution ensures that processing occurs in real-time, allowing for effective analysis and quick response to users. Users can better understand product reviews and make more informed purchasing decisions when sentiment analysis is integrated into the online shopping experience. The proposed approach is a major advance in the development of sentiment analysis for online trading related applications. His ability to conduct real-time sentiment analysis allows him to uncover key insights into customers’ thoughts and preferences. The method provides a viable way for companies to assess consumer satisfaction and sentiment trends by properly classifying product reviews. As a result, companies can ultimately improve the quality of the products and services they provide. Keyword: Online shopping applications, Sentiment analysis, Product reviews, Natural language processing (NLP), support vector machines (SVM), Naive Bayes, Random forest.
... SA employing DL methods has been implemented for different purposes, such as health reviews [124][125][126], financial and product reviews [127,128], services monitoring system reviews [129] and Movie Reviews [21,130,131]. They have also been applied to many languages such as English [132], Chinese [133], Hindi [21], Spanish [134], Lithuanian [135], Arabic [136,137], Bambara-French [138], Persian [139,140], and Malay [141]. ...
Analyzing and understanding the sentiments of social media documents on Twitter, Facebook, and Instagram has become a very important task at present. Analyzing the sentiment of these documents gives meaningful knowledge about the user opinions, which will help understand the overall view on these platforms. The problem of sentiment analysis (SA) can be regarded as a classification problem in which the text is classified as positive, negative, or neutral. This paper aims to give an intensive, but not exhaustive, review of the main concepts of SA and the state-of-the-art techniques; other aims are to make a comparative study of their performances, the main applications of SA as well as the limitations and the future directions for SA. Based on our analysis, researchers have utilized three main approaches for SA, namely lexicon/rules, machine learning (ML), and deep learning (DL). The performance of lexicon/rules-based models typically falls within the range of 55–85%. ML models, on the other hand, generally exhibit performance ranging from 55% to 90%, while DL models tend to achieve higher performance, ranging from 70% to 95%. These ranges are estimated and may be higher or lower depending on various factors, including the quality of the datasets, the chosen model architecture, the preprocessing techniques employed, as well as the quality and coverage of the lexicon utilized. Moreover, to further enhance models’ performance, researchers have delved into the implementation of hybrid models and optimization techniques which have demonstrated an ability to enhance the overall performance of SA models.
... Penelitian oleh A. Feizollah dkk, memanfaatkan kumpulan data ulasan film dengan model DL berhasil mencapai tingkat akurasi sebesar 87,7% pada arsitektur CNN dan 86,64% pada LSTM [9]. Selanjutnya studi oleh A. U. Rehman dkk, mengusulkan model Hybrid CNN-LSTM untuk analisis sentimen pada kumpulan data ulasan film IMDB dan film Amazon mencapai akurasi 91% dibandingkan dengan ML dan model DL tradisional [10]. ...
Sentiment analysis is gaining popularity as the number of internet users increases. Internet users often express their opinions through reviews on websites. The customer opinions expressed have a huge impact on sellers and customer numbers, as many consumers rely on online reviews as a reference when purchasing products. In order to quickly understand sentimental views and tendencies towards a product or event, a text sentiment analysis is performed on the opinions expressed by users. Sentiment analysis focuses on understanding the sentiments contained in the text. One common approach in sentiment analysis is to use Deep Learning (DL) models. This study aims to analyze product sentiment in the Tools & Home category from Amazon using models such as Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). The CNN model is used to extract features from words that reflect short-term sentiment dependencies, while LSTM is used to establish long-term sentiment relationships between words. CNN and LSTM are sophisticated DL models, capable of efficiently processing text data and recognizing relationships and patterns that exist at various levels of abstraction. The purpose of this study is to understand the differences in the performance of the DL model in conducting sentiment analysis, it is hoped that it can also be a reference for those who plan to apply other DL models.
