<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>The theory and technology of artificial intelligence are becoming more and more mature, and its application fields are expanding, greatly changing people’s lifestyles and work modes. The article sorted out the power mechanism and interaction relationship of AI technology to promote the development of e-commerce, and innovated the development mode of e-commerce based on AI empowerment, but also illustrated the ethical challenges brought by its application. Combining the BERT model and LSTM model to establish a sentiment analysis model for online reviews of e-commerce products, using the BiGRU model to obtain users’ long-term preferences, and learning users’ short-term preferences through the multi-head self-attention mechanism to establish a personalized recommendation model for e-commerce products. The study explores the algorithmic discrimination of AI-enabled e-commerce and introduces economic concepts to analyze the consumer’s perceived betrayal and kill-off phenomenon through experiments. At the lowest level, the individualized recommendation model only has an MAE value of 0.679, while the BERT-LSTM model has a sentiment analysis accuracy of 0.837. The average level of perceived betrayal and deliberate attribution of the low-priced group is 4.035 and 4.369, respectively, which is lower than that of the high-price group is lower, and the killing phenomenon brought by algorithmic price discrimination can be alleviated when the elasticity of demand is 0.4.AI-enabled e-commerce needs to be legally regulated from the five perspectives of legislation, law enforcement, judiciary, law-abiding, and law-monitoring as a way of circumventing the algorithmic discrimination and ethical challenges brought about by AI technology.</p>
</abstract>

The theory and technology of artificial intelligence are becoming more and more mature, and its application fields are expanding, greatly changing people’s lifestyles and work modes. The article sorted out the power mechanism and interaction relationship of AI technology to promote the development of e-commerce, and innovated the development mode of e-commerce based on AI empowerment, but also illustrated the ethical challenges brought by its application. Combining the BERT model and LSTM model to establish a sentiment analysis model for online reviews of e-commerce products, using the BiGRU model to obtain users’ long-term preferences, and learning users’ short-term preferences through the multi-head self-attention mechanism to establish a personalized recommendation model for e-commerce products. The study explores the algorithmic discrimination of AI-enabled e-commerce and introduces economic concepts to analyze the consumer’s perceived betrayal and kill-off phenomenon through experiments. At the lowest level, the individualized recommendation model only has an MAE value of 0.679, while the BERT-LSTM model has a sentiment analysis accuracy of 0.837. The average level of perceived betrayal and deliberate attribution of the low-priced group is 4.035 and 4.369, respectively, which is lower than that of the high-price group is lower, and the killing phenomenon brought by algorithmic price discrimination can be alleviated when the elasticity of demand is 0.4.AI-enabled e-commerce needs to be legally regulated from the five perspectives of legislation, law enforcement, judiciary, law-abiding, and law-monitoring as a way of circumventing the algorithmic discrimination and ethical challenges brought about by AI technology.

Applications and Ethical Challenges of Artificial Intelligence Technologies in E-Commerce

Logistics Department of Hunan College of Information

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

The theory and technology of artificial intelligence are becoming more and more mature, and its application fields are expanding, greatly changing people’s...

Applications and Ethical Challenges of Artificial Intelligence Technologies in E-Commerce

Pubblicato online: 05 ago 2024

Pagine: -

Ricevuto: 06 apr 2024

Accettato: 22 giu 2024

DOI: https://doi.org/10.2478/amns-2024-2068

Parole chiave
Bert model, LSTM model, Bigru model, Self-attention mechanism, E-commerce

© 2024 Bin Xiao et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Applications and Ethical Challenges of Artificial Intelligence Technologies in E-Commerce

Pubblicato online: 05 ago 2024

Pagine: -

Ricevuto: 06 apr 2024

Accettato: 22 giu 2024

DOI: https://doi.org/10.2478/amns-2024-2068

Parole chiaveBert model, LSTM model, Bigru model, Self-attention mechanism, E-commerce

© 2024 Bin Xiao et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Parole chiave
Bert model, LSTM model, Bigru model, Self-attention mechanism, E-commerce