Enhancing NEV Brand Equity Through Big Data Analytics: An LDA-LSTM Approach to Mining Online Consumer Reviews

Area of Science:

• * Utilizes advanced big data analytics and artificial intelligence for scientific research in marketing and consumer behavior.
• * Focuses on the automotive industry, specifically the rapidly growing new energy vehicle (NEV) sector.

Background:

• * Intense competition necessitates brand value enhancement for new energy vehicle (NEV) enterprises.
• * Online consumer reviews represent a critical, large-scale data source for understanding brand perception.

Purpose of the Study:

• * To leverage big data analytics on online consumer reviews to improve brand equity for NEV firms.
• * To identify key dimensions of consumer perception and sentiment regarding NEVs.

Main Methods:

• * Collected and processed 5564 online reviews from "Dongche Di" using web scraping and a big data pipeline.
• * Employed word cloud visualization, semantic network analysis, and a Latent Dirichlet Allocation (LDA)-Long Short-Term Memory (LSTM) fusion model for text mining and sentiment analysis.

Main Results:

• * Identified five core NEV brand perception dimensions: range, driving experience, interior space, price, and high-speed performance.
• * Quantified consumer sentiment, revealing prominent negativity in driving experience and overall dominant negative sentiment across reviews.
• * Detected minimal negativity concerning interior space.

Conclusions:

• * Big data analytics effectively scales the understanding of consumer perception for NEV brands.
• * Proposed data-driven brand enhancement strategies based on the Consumer-Based Brand Equity model.
• * Offers a framework for NEV companies to optimize branding through consumer data insights.