A trust-enhanced and preference-aware collaborative method for recommending new energy vehicle

Area of Science:

• Engineering
• Computer Science
• Environmental Science

Background:

• New energy vehicles (NEVs) are crucial for addressing energy scarcity and pollution.
• The abundance of NEV data makes personalized selection challenging for consumers.
• Existing recommendation systems may not fully capture user preferences or social dynamics.

Purpose of the Study:

• To develop a three-stage recommendation method for new energy vehicles (NEVs).
• To integrate user preferences and social trust relationships for improved NEV recommendations.
• To enhance the personalized selection of eco-friendly vehicles.

Main Methods:

• Utilizing the Best-Worst Method (BWM) with hesitant fuzzy preference vectors to determine user criteria weights.
• Calculating demographic similarity and generating trust degrees via NP-IOWA operator for trust-based similarity.
• Constructing a comprehensive user-rating matrix and applying collaborative filtering with trust-based similarity for NEV recommendations.

Main Results:

• The proposed method effectively facilitates personalized NEV recommendations.
• Demonstrated feasibility through a case study.
• Comparative analysis confirmed the advantages of the integrated approach.

Conclusions:

• The three-stage recommendation method enhances the selection of new energy vehicles.
• Incorporating user preferences and social trust significantly improves recommendation accuracy.
• This approach offers a valuable tool for consumers navigating the NEV market.