Sentiment Analysis in E-Commerce Platforms: A Review of Current Techniques and Future Directions

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 90367 - 90382

Published: Jan. 1, 2023

Sentiment analysis (SA), also referred to as opinion mining, has become a widely used real-world application of Natural Language Processing in recent times. Its main goal is identify the hidden emotions behind plain text. SA especially useful e-commerce fields, where comments and reviews often contain wealth valuable business information that great research value. The objective this study examine techniques for current platforms well future directions e-commerce. After examining existing systematic review papers, it was found there lack single comprehensive paper addresses questions. findings can provide researchers field with understanding utilized, insights into directions. Through utilization specific keywords, we have identified 271 papers chosen 54 experimental review. Among these, 26 (representing 48.%) exclusively employed machine Learning techniques, while 24 (44.%) looked addressing through deep learning 4 (7.%) hybrid approach using both techniques. Additionally, our revealed Amazon Twitter emerged two most favored data sources among researchers. Looking ahead, promising avenues include development more universal language models, aspect-based SA, implicit aspect recognition extraction, sarcasm detection, fine-grained sentiment analysis.

Language: Английский

---

Enhancing E-commerce recommendations with sentiment analysis using MLA-EDTCNet and collaborative filtering

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 25, 2025

The rapid growth of e-commerce has made product recommendation systems essential for enhancing customer experience and driving business success. This research proposes an advanced framework that integrates sentiment analysis (SA) collaborative filtering (CF) to improve accuracy user satisfaction. methodology involves feature-level with a multi-step pipeline: data preprocessing, feature extraction using log-term frequency-based modified inverse class frequency (LFMI) algorithm, classification Multi-Layer Attention-based Encoder-Decoder Temporal Convolution Neural Network (MLA-EDTCNet). To address imbalance issues, Modified Conditional Generative Adversarial (MCGAN) generates balanced oversamples. Furthermore, the Ocotillo Optimization Algorithm (OcOA) fine-tunes model parameters ensure optimal performance by balancing exploration exploitation during training. integrated system predicts polarity—positive, negative, or neutral—and combines these insights CF provide personalized recommendations. Extensive experiments conducted on Amazon dataset demonstrate proposed approach outperforms state-of-the-art models in accuracy, precision, recall, F1-score, AUC. By leveraging SA CF, delivers recommendations tailored preferences while engagement highlights potential hybrid deep learning techniques critical challenges systems, including extraction, offering robust solution modern platforms.

Language: Английский

---

Deep Residual Twin directional neural network (DRTNN) based user intention Discovery and product recommendation in e-commerce website

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 127511 - 127511

Published: April 1, 2025

Language: Английский

---

Sentiment Analysis for E-commerce Product Reviews: Current Trends and Future Directions

Published: May 23, 2023

Numerous goods and services are now offered through online platforms due to the recent growth of transactions like e-commerce. Users have trouble locating product that best suits them from numerous products available in shopping. Many studies deep learning-based recommender systems (RSs) focused on intricate relationships between attributes users items. Deep learning techniques used consumer or item-related traits improve quality personalized many areas, such as tourism, news, Various companies, primarily e-commerce, utilize sentiment analysis enhance effectively navigate today's business environment. Customer feedback regarding a is gathered analysis, which uses contextual data split it into separate polarities. The explosive rise e-commerce industry has resulted large body literature different perspectives. Researchers made an effort categorize recommended future possibilities for study field grown. There several challenges fake reviews, frequency user advertisement click fraud, code-mixing. In this review, we introduce overview preliminary design Second, concept learning, discussed. Third, represent versions commercial dataset. Finally, explain various difficulties facing RS research directions.

Language: Английский

---

A Hybrid Deep Learning Framework for Efficient Sentiment Analysis

International Journal of Advanced Computer Science and Applications, Journal Year: 2023, Volume and Issue: 14(12)

Published: Jan. 1, 2023

In the era of Microblogging and rapid growth online platforms, an exponential rise is shown in volume data generated by internet users across various domains. Additionally, creation digital or textual expanding significantly. This because consumers respond to comments made on social media platforms regarding events products based their personal experiences. Sentiment analysis usually used accomplish this kind classification a large scale. It described as process going through all user reviews that are discovered product reviews, events, similar sources order look for unstructured text comments. Our study examines how deep learning models like LSTM, GRU, CNN, hybrid (LSTM+CNN, LSTM+GRU, GRU+CNN) capture complex sentiment patterns data. we integrating BOW TF-IDF complementing features improve model predictive power. CNN with RNNs consistently improves outcomes, demonstrating synergy between convolutional recurrent neural network architectures recognizing nuanced emotion subtleties.In addition, typically outperforms enhancing accuracy.

Language: Английский

---

Pengaruh E-commerce dan Kemudahan Transaksi Terhadap Perubahan Pola Konsumsi Dalam Era Digital Di Indonesia

Deleted Journal, Journal Year: 2023, Volume and Issue: 1(01), P. 32 - 42

Published: Dec. 30, 2023

Studi ini meneliti bagaimana perubahan kebiasaan konsumsi di era digital dipengaruhi oleh e-commerce dan kemudahan transaksi Indonesia. Untuk menyelidiki hubungan antara tiga konstruk utama - E-commerce, Kemudahan Transaksi, Pola Konsumsi digunakan analisis kuantitatif dengan menggunakan pemodelan persamaan struktural. Sampel sebanyak 150 orang diikutsertakan dalam penelitian ini, data dilakukan metode statistik yang canggih seperti SEM-PLS 4. Temuan menunjukkan bahwa terdapat korelasi kuat menguntungkan Transaksi Konsumsi, serta E-commerce Konsumsi. Hasil pola pembelian konsumen Indonesia sebagian besar disebabkan semakin populernya perdagangan pengalaman lebih baik. Kesesuaian model struktural disarankan dikonfirmasi sejumlah indeks kecocokan nilai R-Square. menambah pengetahuan kita tentang berperilaku memberikan informasi berguna bagi perusahaan pengambil keputusan berusaha memahami mengambil keuntungan dari tren Makalah juga menyoroti kekurangannya merekomendasikan arah untuk masa depan memperdalam pemahaman fenomena dinamis ini.

---

The Influence of Digital Technology Adoption by Small and Medium Enterprises and Online Consumer Behavior on the Success of E-commerce Platforms in Bandung City

West Science Journal Economic and Entrepreneurship, Journal Year: 2023, Volume and Issue: 1(04), P. 153 - 159

Published: April 30, 2023

This study looks into the complex relationships that exist between use of digital technology by Small and Medium Enterprises (SMEs) in Bandung City, online consumer behavior, performance e-commerce platforms. Through structured surveys a quantitative approach, information about integration, customer behavior patterns, markers success was gathered from 121 SMEs. The results show dynamics are diverse, there is moderate level acceptance technology, both characteristics have considerable beneficial impact on e-commerce. key component emerges trust, underscoring significance building confidence economy. adds to our knowledge interactions occur engagement adoption, providing SMEs navigating landscape with useful insights.

Language: Английский

---