Balancing consumer and business value of recommender systems: A simulation-based analysis

Automated recommendations can nowadays be found on many e-commerce platforms, and such create substantial value for consumers providers. Often, however, not all recommendable items have the same profit margin, providers might thus tempted to promote that maximize their profit. In short run, accept non-optimal recommendations, but they may lose trust in long run. Ultimately, this leads problem of designing balanced recommendation strategies, which consider both consumer provider lead sustained business success. This work proposes a simulation framework based agent-based modeling designed help explore longitudinal dynamics different strategies. our model, agents receive from providers, perceived quality influences consumers’ over time. We design several strategies either give more weight or utility. Our simulations show hybrid strategy puts utility without ignoring profitability considerations highest cumulative results increase about 20% compared pure oriented also find social media reinforce observed phenomena. case when heavily rely media, best further increases. To ensure reproducibility foster future research, we publicly share flexible framework.