A New Multi-Attribute Decision Making Method for Overvalued Star Ratings Adjustment and Its Application in New Energy Vehicle Selection

Abstract

Under the global consensus of carbon peaking and carbon neutrality, new energy vehicles have gradually become mainstream, driven by the dual crises regarding the atmospheric environment and energy security. When choosing new energy vehicles, consumers prefer to browse the post-purchase reviews and star ratings of various new energy vehicles on platforms such as DCar.com. However, it is easy for consumers to become lost in the high-star text reviews and mismatched reviews. To solve the above two issues, this study selected nine new energy vehicles and used a multi-attribute decision making method to rank the vehicles. We first designed adjustment rules based on star ratings and text reviews to cope with the issue of high star ratings but negative text reviews. Secondly, we classified consumers and recommended the optimal alternative for each type of consumer to deal with the issue of mismatched demands between review writers and viewers. Finally, this study compared the ranking results with the sales charts of the past year to verify the feasibility of the proposed method initially. The feasibility and stability of the proposed method were further verified through comparative and sensitivity analyses.

Machine Learning Based Risk Management of Credit Sales in Small and Mid-Size Business

“…The study defined the pandemic's distortion of social and business features, among which were remote work, digitalization, resistance of flows of supply, and shifts of consumer behavior. [23] Pairs of Yu, Zhang, Du, and Chen (2023) make the decision model with multiple attributes that can be used in the multi-attribute, as well as add, revise, or Their research contributed to the field of decision science by introducing a novel approach to address rating biases and enhance decision-making processes. The range of stuides conducting the business and technology aspects as well as addressing some pertain issues is the line that the literature review has exposed.…”