Opinion Mining-Based Conjoint Analysis of Consumer Brands

Nowadays, online customer reviews in the form of feedback and ratings are available for a large number of product or service-based categories. These product reviews signify individually perceived strengths and shortcomings of a given product category, which can essentially provide design engineers and market researchers with invaluable insights into product design. In order to analyze the available individual customer opinions and turn them into aggregate consumer preferences that can be used in the product development and improvement process, we proposed a hybrid of sentiment analysis and conjoint analysis to determine the best brand among a list of brands for a specified product category. The outcome of this hybrid model is compared with the hybrid of sentiment analysis and multi-criteria decision making. The proposed framework estimates the relative effect of product attributes and brand names on the overall preference of consumers. We demonstrated the effectiveness of our approach using product review data taken from the literature. Our proposed approach turned out to be useful in determining the best of market availability.