Terms Of Frequency Research Articles

A pediatric emergency prediction model using natural language process in the pediatric emergency department

This study developed a predictive model using deep learning (DL) and natural language processing (NLP) to identify emergency cases in pediatric emergency departments. It analyzed 87,759 pediatric cases from a South Korean tertiary hospital (2012–2021) using electronic medical records. Various NLP models, including four machine learning (ML) models with Term Frequency-Inverse Document Frequency (TF-IDF) and two DL models based on the KM-BERT framework, were trained to differentiate emergency cases using clinician transcripts. Gradient Boosting, among the ML models, performed best with an AUROC of 0.715, AUPRC of 0.778, and F1-score of 0.677. DL models, especially the fine-tuned KM-BERT model, showed superior performance, achieving an AUROC of 0.839, AUPRC of 0.879, and F1-score of 0.773. Shapley-based explanations provided insights into model predictions, underlining the potential of these technologies in medical decision-making. This study demonstrates the potential of advanced DL techniques for NLP in emergency medical settings, offering a more precise and efficient approach to managing healthcare resources and improving patient outcomes.

AI-Powered Keyword Extraction System Using NLP Techniques for Contextual Insights and Document Accessibility

This study presents the development of an AI-powered keyword extraction system using Python, aimed at efficiently identifying significant keywords from unstructured text data. Leveraging advanced natural language processing (NLP) techniques, the system integrates multiple methodologies, including Term Frequency-Inverse Document Frequency (TF-IDF), TextRank, Latent Semantic Analysis (LSA), and Part-of-Speech (POS) tagging. The incorporation of POS tagging enhances the accuracy of keyword identification by focusing on essential parts of speech, such as nouns and verbs, thereby filtering out irrelevant terms and ensuring the extraction of contextually meaningful keywords. To enhance user experience, the system includes innovative features such as a mobile document transfer facility that allows users to transfer documents via QR code and an option to download the extracted keywords as a document or PDF. These functionalities are designed to streamline accessibility and usability for diverse user groups. Performance evaluation was conducted using a dataset of research articles, demonstrating the system's effectiveness in accurately reflecting document content through keyword extraction. The results highlight its potential applications across various domains, including academia, marketing, and data science. This project offers a robust, intuitive, and user-friendly solution to make complex text data more accessible and actionable, contributing to the growing demand for efficient keyword extraction tools. Keywords- ai-powered keyword extraction, natural language processing (nlp), term frequency- inverse document frequency (tf-idf), textrank algorithm, latent semantic analysis (lsa), part-of- speech (pos) tagging

Spam and Phishing Whatsapp Message Filtering Application Using TF - IDF and Machine Learning Methods

The rapid development of communication technology has led to an increase in the number of unwanted messages, such as spam and phishing attempts. However, this progress has not been accompanied by sufficient user awareness of the basics of technology use. Additionally, the enforcement of laws regarding internet-based crimes remains unclear, further increasing the risk for users of internet technology to fall victim to such crimes. As one of the media prone to spam and phishing, WhatsApp is the focus of this research, which aims to develop an application capable of filtering spam and phishing messages. The application employs the TF-IDF (Term Frequency-Inverse Document Frequency) method and machine learning using the Random Forest model. It is developed using the MVVM (Model-View-ViewModel) architecture, enabling the separation of business logic from the user interface, thereby improving development and maintenance efficiency. The research findings demonstrate that the combination of TF-IDF and Random Forest achieves high accuracy in classifying spam and phishing messages. Performance evaluation using a confusion matrix reveals an accuracy rate of 92%. For the safe message class, the precision, recall, and F1 scores are 89%, 95%, and 92%, respectively, while for the dangerous message class, the scores are 95%, 88%, and 92%, respectively. Furthermore, the integration of the model and application performed exceptionally well, as evidenced by black-box testing results. All test scenarios were met, successfully detecting test messages with 98% accuracy. Therefore, the developed application provides enhanced protection for WhatsApp users against digital threats.

Explainable Predictive Model for Suicidal Ideation During COVID-19: Social Media Discourse Study.

Studying the impact of COVID-19 on mental health is both compelling and imperative for the health care system's preparedness development. Discovering how pandemic conditions and governmental strategies and measures have impacted mental health is a challenging task. Mental health issues, such as depression and suicidal tendency, are traditionally explored through psychological battery tests and clinical procedures. To address the stigma associated with mental illness, social media is used to examine language patterns in posts related to suicide. This strategy enhances the comprehension and interpretation of suicidal ideation. Despite easy expression via social media, suicidal thoughts remain sensitive and complex to comprehend and detect. Suicidal ideation captures the new suicidal statements used during the COVID-19 pandemic that represents a different context of expressions. In this study, our aim was to detect suicidal ideation by mining textual content extracted from social media by leveraging state-of-the-art natural language processing (NLP) techniques. The work was divided into 2 major phases, one to classify suicidal ideation posts and the other to extract factors that cause suicidal ideation. We proposed a hybrid deep learning-based neural network approach (Bidirectional Encoder Representations from Transformers [BERT]+convolutional neural network [CNN]+long short-term memory [LSTM]) to classify suicidal and nonsuicidal posts. Two state-of-the-art deep learning approaches (CNN and LSTM) were combined based on features (terms) selected from term frequency-inverse document frequency (TF-IDF), Word2vec, and BERT. Explainable artificial intelligence (XAI) was used to extract key factors that contribute to suicidal ideation in order to provide a reliable and sustainable solution. Of 348,110 records, 3154 (0.9%) were selected, resulting in 1338 (42.4%) suicidal and 1816 (57.6%) nonsuicidal instances. The CNN+LSTM+BERT model achieved superior performance, with a precision of 94%, a recall of 95%, an F1-score of 94%, and an accuracy of 93.65%. Considering the dynamic nature of suicidal behavior posts, we proposed a fused architecture that captures both localized and generalized contextual information that is important for understanding the language patterns and predict the evolution of suicidal ideation over time. According to Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP) XAI algorithms, there was a drift in the features during and before COVID-19. Due to the COVID-19 pandemic, new features have been added, which leads to suicidal tendencies. In the future, strategies need to be developed to combat this deadly disease.

Spectral Clustering-Based Particle Swarm Optimization Algorithm for Document Clustering

The process of automatically grouping documents into clusters such that the documents in one cluster are very comparable to the documents in the remaining clusters have been known as document clustering. Due to its broad application in a number of fields, including search engines, web mining, and information retrieval, it has been the subject of much research. It involves clustering documents that are identical to one another and calculating how identical they are. It facilitates simple navigation by offering effective document representation as well as visualization. Hence, this research paper plans to perform the document clustering using the nature inspired optimization technique. Initially, the dataset is manually gathered from different sources. Next, the data preparation has been done for extracting the text content from the published documents. These prepared data undergo pre-processing for removing the punctuations, stop words, and lowercase conversion. The features are extracted from these pre-processed data utilizing the Term Frequency-Inverse Document Frequency (TF-IDF) approach for extracting the keywords. The extracted features undergo the final clustering phase employing the spectral clustering algorithm, in which the parameter tuning has been done by the nature inspired optimization algorithm referred as Particle Swarm Optimization (PSO) with the consideration of silhouette score maximization as the objective function. This proposed spectral clustering-PSO clusters the final output into six classes such as data mining, deep learning, image, machine learning, network, and sports respectively. The proposed document clustering model describes its betterment over the remaining techniques with respect to distinct measures. The proposed spectral clustering-PSO in terms of silhouette score is 51.92%, 70.81%, 45.93%, and 20.89% better than JA-GWO, tpLDA, HDMA, and Net2Vec respectively. Similarly, the proposed spectral clustering-PSO in terms of davies bouldin score is 89.69%, 58.48%, 32.67%, and 13.99% advanced than JA-GWO, tpLDA, HDMA, and Net2Vec respectively.

Bridging BI and AI Enhancing Operational Efficiency in the Chinese Financial Sector

This study explores the impact of Business Intelligence (BI) systems on operational efficiency (OE) and the transition to Artificial Intelligence (AI) technologies in financial firms from Shanghai and Shenzhen stock markets (2007-2021). It investigates whether existing BI platforms underpin AI adoption, using Data Envelopment Analysis as a proxy for OE and the frequency of terms like Big Data and data mining in annual reports to indicate BI usage. Employing Heckman's two-stage and Hausman firm fixed effect model addresses potential endogeneity. Results show significant OE improvements post-BI adoption, with increasing benefits over time and enhanced by R&D intensity. Additionally, this research extends to global information management, linking BI capabilities with AI readiness and offering insights into strategic technology management in the financial sector, aligning with shifts towards AI in business, thereby impacting local and global information strategies.

Performance Comparison of Text Weighting Schemas on NMF-Based Topic Analysis

Nowadays, it is feasible to analyze text data that is being generated at an exponential rate by transforming it into a sparse matrix of big size using a certain weighting method. A comprehensive text weighting approach consists of three fundamental components: Term Frequency, Document Frequency, and Vector Normalization. The multiplication of these three components yields numerical values that indicate the significance of a word for a text. Nevertheless, the unprocessed state of these values is unsuitable for the semantic analysis of textual material. There are multiple techniques available for this objective, and Topic Analysis, which seeks to identify subjects discussed in extensive text collections, is one of these techniques. The Non-Negative Matrix Factorization (NMF) approach is commonly employed in topic analysis. It involves transforming an input matrix into the product of two or more matrices, using both random and deterministic beginning values. This study involved conducting tests on a dataset of 20,000 articles sourced from Wikipedia, the online encyclopedia, with the aim of investigating the impact of text weighting methods and initial value approaches commonly employed in the literature on the NMF method. The number of clusters to be used in the studies was determined using an analytical procedure, which employed an upper limit. The results indicate that the “lnc” and “nnc” weighting schemes yielded the highest performance in NMF. These findings demonstrate that employing the “lnc” or “nnc” weighting scheme will lead to more favorable outcomes in the domain of topic analysis.

What do citizens think of central bank digital currencies in the euro context? A Twitter sentiment analysis

PurposeThis paper delves into the discourse surrounding central bank digital currencies (CBDC) from the perspective of citizens’ views expressed on Twitter, on the potential implementation of a digital euro. The paper sheds light on citizens’ sentiments regarding CBDCs in relation to the euro and the euro area.Design/methodology/approachUsing “TextBlob” sentiment analysis, this paper analyzes 4,462 Tweets containing the keywords “cbdc” and “euro” posted between September 14, 2018, and June 19, 2023. It explores citizens’ perceptions and concerns, as well as the general sentiment conveyed by the Tweets, through sentiment analysis and word-cloud generation. Text analysis and the “KMeans algorithm” are used to identify similar groups of Tweets. Statistical analysis of Term Frequency-Inverse Document Frequency (TF-IDF) keyword values in each cluster provides information about the relative distribution of words within clusters. In a further step, sentiment analysis is performed on each cluster by rating it positively, neutrally or negatively to identify the relevant dominant themes.FindingsThis paper reveals the evolving nature of CBDC-related discourse on Twitter over a four-year period, demonstrating a slight overall trend of positive sentiments. The distribution includes Tweets with a pronounced negative and positive sentiment, highlighting the need for clear and citizen-friendly communication through social media channels (such as Twitter or LinkedIn) in addition to the traditional channels used by official institutions.Research limitations/implicationsThe extraction of Tweets was limited to English; bilingual Tweets containing English words were also considered. Future research could be expanded to include Tweets in additional languages and compare them to English Tweets. Geographical data on the origin of the Tweets were not considered due to potential inaccuracies resulting from the use of VPNs.Practical implicationsThe findings highlight citizens’ views, expectations and perceptions on the digital euro as a type of CBDC, suggesting ways in which the European Central Bank (ECB) can ensure customer satisfaction with its successful implementation.Originality/valueThe originality of the article revolves around the focus on the digital euro and on analyzing European citizens’ opinions regarding its implementation. In contrast to previous literature, this research discusses critically the pros and cons and possible design aspects of the new digital euro, by taking a citizen-centric approach and focusing solely on one type of currency. Technological advances and the digital era have left their mark on how society communicates and behaves. The implementation of CBDCs will affect the digital society. Therefore, the paper sparks a debate about European citizens’ sentiments toward adopting the digital euro, as a new digital currency, which will impact their lifestyle and financial decisions.

Optimization of English Grammar Error Automatic Detection Algorithm Based on Natural Language Processing

A text’s grammatical errors (GEs) are passages that are broken down to accept grammar standards. Resolving grammatical faults and inconsistencies in a document is the aim of GEs. Different strategies can focus on varying textual spans, from individual phrases to whole documents. Common GEs include improper word choice, punctuation, and syntax. The development of natural language processing (NLP) has fundamentally changed the way textual data is analyzed and processed, leading to notable advancements in automated grammatical mistake detection systems. In this study, we proposed a novel starling murmuration-optimized dense recurrent neural network (SMO-DRNN) model for the detection of English grammar errors. In this study, we collected text samples that were analyzed by marking different types of GEs. Automatic reading also involves converting textual data from English compositions into numerical values for further computation. Data pre-processing techniques include tokenization, stop word removal, stemming, and lemmatization. To extract relevant information from the pre-processing data, term frequency-inverse document frequency (TF-IDF) feature extraction was used for accurate grammar detection. The proposed approach is compared to the traditional algorithms. The overall results show that the proposed approach performed better than the existing method in terms of accuracy and loss, ROC (0.95), recall (92%), precision (96%), and F-0.5 score (94%), for identifying English grammar errors. The suggested method effectively combines current NLP strategies to offer a highly accurate method for identifying grammar mistakes in English.

Integrating Machine Learning and Genetic Algorithms to Enhance Gene-Disease Classification: An XBNet-Based Framework

In bioinformatics, the classification of gene-disease associations is crucial. It directly affects whether we can untangle the genetic roots of various disease as well as if we will find some justifiable therapy for these cured diseases.Using XBNet to construct genetic algorithms for higher accuracy and speeds of gene-disease classification--this is the method developed in the book.Consisting of gene expression profiles for six diseases--Alzheimer's, Asthma, Cancer, Diabetes, Fabry and Down syndrome--our research has applied a comprehensive pre-processing technique to this data set from Kaggle. This has included such things as eliminating stop-words and punctuation marks and tokenization. Using the terms of Frequency (TF) and of Term Frequency-Inverse Document Frequency (TF-IDF method) for features extraction, our text data on genes are transformed into numerical axes fit for input to machine learning models.

Mediating effects of social support between telework and well-being: protocol for a systematic review

IntroductionResearch on telework has grown dramatically in recent years, leading to a fragmented set of findings on its effect on workers’ well-being. One of the most studied social mechanisms concerns...

An efficient smart phone application for wheat crop diseases detection using advanced machine learning.

Globally, agriculture holds significant importance for human food, economic activities, and employment opportunities. Wheat stands out as the most cultivated crop in the farming sector; however, its annual production faces considerable challenges from various diseases. Timely and accurate identification of these wheat plant diseases is crucial to mitigate damage and enhance overall yield. Pakistan stands among the leading crop producers due to favorable weather and rich soil for production. However, traditional agricultural practices persist, and there is insufficient emphasis on leveraging technology. A significant challenge faced by the agriculture sector, particularly in countries like Pakistan, is the untimely and inefficient diagnosis of crop diseases. Existing methods for disease identification often result in inaccuracies and inefficiencies, leading to reduced productivity. This study proposes an efficient application for wheat crop disease diagnosis, adaptable for both mobile devices and computer systems as the primary decision-making engine. The application utilizes sophisticated machine learning techniques, including Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and AdaBoost, combined with feature extraction methods such as Count Vectorization (CV) and Term Frequency-Inverse Document Frequency (TF-IDF). These advanced methods collectively achieve up to 99% accuracy in diagnosing 14 key wheat diseases, representing a significant improvement over traditional approaches. The application provides a practical decision-making tool for farmers and agricultural experts in Pakistan, offering precise disease diagnostics and management recommendations. By integrating these cutting-edge techniques, the system advances agricultural technology, enhancing disease detection and supporting increased wheat production, thus contributing valuable innovations to both the field of machine learning and agricultural practices.

IDEA-6G: Revolutionizing 6G Networks with Integrated Digital Twin and Self-Healing Mechanisms

6G network is an innovative concept of connectivity, which offers unparalleled speeds, ultralow latency, and extensive device connectivity that surpass the capabilities of the current 5G networks. However, challenges such as network congestion and security threats pose significant hurdles to ensuring reliable and stable network performance. A novel Integrated Digital twin and self-healing mechanisms for 6G networks (IDEA-6G) approach has been proposed for addressing these challenges and for the security and performance of the 6G network. The proposed method leverages the Digital Twin (DT) sub-layer to bridge the physical and digital worlds, enabling real-time synchronization and monitoring of network assets. Meticulous feature extraction using Term Frequency - Inverse Document Frequency (TF-IDF) techniques and the innovative Generative Adversarial Networks and Long Short-Term Memory (GAN-LSTM) model have helped the approach in the enhancement of security monitoring capabilities and efficient detection of cyber-attacks within virtual models. Additionally, Deep Neural Networks (DNNs) facilitate informed decision-making for effective self-healing actions in response to identified threats. The effectiveness of the proposed IDEA-6G is compared with the existing B5GEMINI, DTFV, and DITEN techniques. Results of the proposed IDEA-6G technique indicate superior performance prediction accuracy, detection rates, load balancing, and service delay reduction. The detection of the proposed IDEA-6G technique is 17.55%, 27.54%, and 7.38% higher than the existing B5GEMINI, DTFV, and DITEN techniques respectively.

Automatic Youtube Comment Spam Detection using Navie Bayes and Logistic Regression

During recent years, the volume and relevance of user-generated content on platforms such as YouTube have created a predicament in spam comment detection and filtering. This paper proposes an automated system for spam comment detection using Naive Bayes and Logistic Regression algorithms. After preprocessing, the spam comments go through tokenization, stop-word removal and stemming, while feature extraction is done using TF-IDF (Term Frequency-Inverse Document Frequency), which translates the comments into a numerical representation. Labeled datasets of YouTube comments introduce spam or non-spam categories for Naive Bayes and Logistic Regression model training. Performance is evaluated using generalized metrics like accuracy, precision, recall, and F1-score. Both models detect spam effectively, while Logistic Regression performs at a clasp higher level than Naive Bayes, (Support vector machine),KNN(k-nearest neighbors). The resulting machine might belong to YouTube's comment moderation scheme, filtering spam automatically in real-time, optimizing user experience to engage in content-rich communication as against annoying content. Key Words: Naïve Bayes, Logistic Regression, TF-IDF, Youtube Comments

Applications of machine learning and artificial intelligence in the oil and gas industry: a study of keywords and research results

The integration of advanced machine learning (ML) and artificial intelligence (AI) techniques in the oil and gas industry is rapidly evolving, focusing on improving operational predictions and optimizations. This study investigates the frequency and interconnectedness of key terms associated with these technologies through network diagrams and academic research results. An extensive search in the Scopus database revealed significant trends and patterns, highlighting the central role of forecasting and optimization within the industry. The interdisciplinary nature of the research underscores the need for collaboration across various fields to tackle complex challenges. While data quality and infrastructure pose challenges, the potential for enhanced efficiency, reduced costs, and increased safety through AI and ML applications is substantial. Case studies demonstrate the practical benefits, and future advancements promise deeper integration of these technologies into industry operations.

NaturePred: A Tool for Revolutionizing Natural Product Classification with Artificial Intelligence

Background: The identification and classification of natural products are vital in drug discovery and bioactive compound exploration. Traditional methods are laborious and timeconsuming, necessitating innovative tools for accurate predictions using advanced AI techniques. Objectives: This paper presents NaturePred, a user-friendly tool designed to predict the class of natural products and calculate eight physicochemical properties of protein sequences. It aims to accurately predict five distinct classes of natural product biosynthetic gene clusters (BGCs): Polyketide Synthases (PKS), Non-ribosomal Peptide Synthetases (NRPS), Ribosomally Synthesized and Post- Translationally Modified Peptides (RiPPs), Terpenes, and PKS-NRPS Hybrids. It also addresses reliability in multi-class classification with a 90% confidence score threshold. Method: NaturePred offers three input options: single protein sequence, CSV file, or GenBank (.gbk) file. It uses a pipeline with a Natural Language Processing model based on TF-IDF (Term Frequency- Inverse Document Frequency) and a Logistic Regression classifier. Predictions are made if the confidence score exceeds 90%; otherwise, "None of the above class" is predicted. Evaluation with unseen data from the MiBIG database shows high accuracy (~96%) in assigning BGCs. Results: NaturePred provides accurate predictions with high confidence scores, demonstrating reliability across different datasets. It calculates eight physicochemical properties of protein sequences, offering valuable insights for further analysis. Conclusion: NaturePred's integrated features, including versatile input options, accurate predictions, and physicochemical property calculations, make it an indispensable tool in natural product research. By addressing classification challenges, NaturePred facilitates drug discovery and bioactive compound exploration, advancing the field. Tool available: (http://login1.cabgrid.res.in:5101/).

Integrating Neutrosophic Vague N-Soft Sets with Chimp Optimization Algorithm for Sentiment Analysis on Social Media

The swift development in social media through the internet produces vast data in a real-time scenario that has startling effects on large datasets. It generated the high-level use of sentiments and emotions in social networking media. Sentiment analysis (SA) using a neutrosophic set presents a new technique to handle the integral ambiguity and uncertainty in text datasets. Different from classical approaches, which categorize sentiment as positive, negative, or neutral, the neutrosophic set allows for the comparison analysis of truth-, indeterminacy-, and falsie-membership functions for all the sentiments. This allows a more flexible and nuanced representation of sentiments, which accommodates the contradictions and complexities commonly depicted in natural language. SA can accomplish high performance and depth in interpreting and understanding the emotions expressed in uncertain and diverse text datasets by leveraging a neutrosophic set. This manuscript presents a Neutrosophic Vague N-Soft set with a Chimp Optimization Algorithm for Sentiment Analysis (NVNSS-COASA) technique on Social Media. The NVNSS-COASA technique is initiated by the comprehensive preprocessing stage to normalize and clean the text dataset, which ensures superior input for the succeeding stage. Then, the Term Frequency-Inverse Document Frequency (TF-IDF) mechanism is employed to convert the preprocessed text into mathematical features, which capture the word importance in terms of datasets. Subsequently, a strong NVNSS classifier is employed for accurately categorizing the sentiment. We integrate COA for the parameter tuning to further improve the performance of the method. The simulation outcomes emphasized that the NVNSS-COASA method shows superior outcomes over other techniques. The outcomes indicated that the NVNSS-COASA can able to deliver reliable and precise insights from the text dataset.

Text vectorization in sentiment analysis: A comparative study of TF-IDF and Word2Vec from Amazon Fine Food Reviews

Sentiment analysis is a practical tool for marketing and branding teams. Companies can collect and analyze opinions or reviews from social media platforms, blog posts, and other numerous forums. It may help them acquire positive feedback to reinforce strengths or identify negative emotions to make improvements. The research is to compare two text vectorization methods in opinion mining: Term Frequency-Inverse Document Frequency (TF-IDF) and Word2Vec, using Amazon Fine Food Reviews dataset. This study will use these two methods to vectorize preprocessed text data and also input the vectorized data to the emotion classification model, analyzing the performance of two methods in the emotion classification task. The consequence indicates that the former outperforms the latter in handling large datasets, particularly in distinguishing between different sentiment categories, but latter is superior in capturing the semantic relationship of words. Therefore, it is suggested that the advantages of the two methods be combined in practical applications to improve the accuracy and efficiency.

A HYBRID APPROACH FOR EMOTION-DRIVEN GAME RECOMMENDATIONS USING TEXT, VOICE AND IMAGE RECOGNITION

This implementation presents a game recommendation system that utilizes natural language processing (NLP) techniques to provide personalized game suggestions based on user preferences. The system processes a dataset of games containing descriptions and emotional tones to determine relevant recommendations. It employs TF-IDF (Term Frequency-Inverse Document Frequency) vectorization to transform textual data into numerical representations, enabling meaningful comparisons between game content and user input. The cosine similarity metric is then used to assess the closeness of games to the given preferences. The recommendation process begins by filtering the dataset based on the user's specified emotional tone, ensuring that only games matching the desired sentiment are considered. Next, the input content is vectorized, and similarity scores are computed against all games in the dataset. The system ranks the games based on similarity and retrieves the top five matches. This approach enhances game discovery by allowing users to find games that align with both their emotional state and gameplay preferences, offering a more personalized gaming experience.

Empirical Analysis of the State-of-the-Art Models for Handling Polarity Shifts Due to Implicit Negation in Mobile Phone Reviews

Purpose–This paper presents a comprehensive empirical analysis focusing on sentiment flux within state-of-the-art models designed for handling polarity shifts due to implicit negation in Amazon mobile phones' reviews. Method–The research evaluates diverse models across five categories: traditional machine learning (ML), deep learning (DL), and hybrid models combining both approaches. Various feature extraction, feature selection, and data augmentation techniques are tested on Amazon mobile phone reviews dataset. BERT and LSTM are used for deep learning while SVM and Naive Bayes are used for traditional ML. ANOVA is used to identify the presence or absence of significant differences and interactions among these entities. Results –DL shows superior performance compared to traditional ML models. ANOVA analysis shows significant performance differences between conventional ML and DL models. Traditional ML models interact significantly with feature extraction and selection techniques while DL models do not. Traditional ML models do not interact significantly with data augmentation methods while DL models do. FastText extraction outperforms word2vec; Back translation outperforms synonym replacement while recursive feature selection (RFE) surpasses TF-IDF (Term Frequency-Inverse Document Frequency). The BERT and LSTM exhibit one of the strongest performances. Conclusion –The study concludes that DL models are more effective. Data augmentation techniques significantly impact the performance of DL models, with back translation showing superior performance over synonym replacement. This provides a leverage point in developing an improved model in the future. Recommendations –Future research should focus on developing a hybrid model for Enhanced Polarity Shift Management of Mobile Phone Reviews using Contextual Back Translation Augmented by Seq2seq Perturbations. This aims at leveraging contextual back translation and Seq2seq perturbations to generate a diverse interpretation that consequently improves the model's ability to handle nuanced expressions of sentiments due to implicit negation with enhanced accuracy, generalizability, robustness to polarity shifts, and contextual understanding. Research Implications –The findings provide valuable insights into the development of state-of-the-art models, offering a promising direction for further research in sentiment analysis. Keywords –empirical analysis, hybrid, perturbations, implicit negation, sentiment flux