Social Media-Driven E-Commerce Growth Through Advanced Technologies

Social media-driven e-commerce has emerged as a dominant force in the digital marketplace, fueled by the integration of advanced technologies. Adoption of AI, machine learning, AR, and big data analytics-which have transformed company interactions with customers-is mostly responsible for this expansion. By using consumer behavior research, AI and ML provide individualized marketing methods that boost conversion rates and user experience. Augmented reality may allow shoppers to virtually test things before buying, improving their shopping experience. Big data analytics may also explore customer preferences and market trends to guide inventory management and targeted advertising. Blockchain technology and other secure payment methods enable transparent, trustworthy transactions. The following abstract discusses how social media e-commerce's technology advances may boost user engagement, simplify operations, and boost global digital economy development. The suggested technique outperforms existing methods in precision, accuracy, recall, F1-score, and user engagement rates. It succeeds when Random Forest (88.5% accuracy), Logistic Regression (85.0% accuracy), and SVM (90%) fail. With <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 0 \%}$</tex> accuracy, the model exceeds logistic regression, which only achieves 80 % accuracy. Its <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 4. 0 \%}$</tex> recall rate suggests it can identify more fraudulent transactions. The model's ability to maintain both accuracy and recall is shown by its F1-score of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 1. 0 \%}$</tex>. With <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{8 8. 0 \%}$</tex>, user involvement rates show the system's success. Recorded at 150 ms, transaction speed is noticeably lower than logistic regression (LR) and other traditional methods. User satisfaction is also high, with an overall satisfaction rate of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 5. 0 \%}$</tex>, compared to only <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{7 5. 0 \%}$</tex> for LR. The model's ease of use is <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 2. 0} \%$</tex>, higher than Logistic Regression's 70.0%. The likelihood to recommend the system is <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 3. 0 \%}$</tex>, indicating strong endorsement from users.