Summary

Recommender systems have their roots in various research areas, such as information retrieval, information filtering, and text classification, and apply methods from different fields, such as machine learning, data mining, and knowledge-based systems. With this book, we aimed to give the reader a broad overview and introduction to the area and to address the following main topics:

• Basic recommendation algorithms: We discussed collaborative and content-based filtering as the most popular recommendation technologies. In addition, the basic recommendation schemes, as well as different optimizations, limitations, and recent approaches, were presented.

• Knowledge-based and hybrid approaches: As the value of exploiting additional domain knowledge (in addition to user ratings or item “content”) for improving a recommender system's accuracy is undisputed, two chapters were devoted to knowledge-based and hybrid recommender systems. We discussed both knowledge-based recommendation schemes, such as constraint and utility-based recommendation, as well as possible hybridization strategies.

• Evaluation of recommender systems and their business value: In most cases, recommender systems are e-commerce applications. As such, their business value and their impact on the user's decision-making and purchasing behavior must be analyzed. Therefore, this book summarized the standard approaches and metrics for determining the predictive accuracy of such systems in the chapter on recommender systems evaluation. A further chapter was devoted to the question of how recommender systems can influence the decision-making processes of online users. Finally, a comprehensive case study demonstrated that recommender systems can help to measurably increase sales.