A Product Generation Algorithm for Revenue and Consumer Rating Optimization

Abstract

Consumer-facing companies seek to develop brand loyalty and increase their market share by providing the best priced and highest quality products. Some industries have started using consumer review data to power recommendation algorithms that direct consumers toward existing products they may enjoy. In this thesis, we expand upon this idea by developing a process that uses consumer preferences to come up with ideas for new products to create. This thesis outlines the idea for a product generation algorithm that identifies important product features for different consumer groups and generates proposals for products that will have a high rating and generate high sales.