Evaluating link prediction: New perspectives and recommendations
This work addresses the need for more rigorous and controlled evaluation in link prediction research, offering incremental improvements to standard practices.
The paper tackles the problem of evaluating link prediction methods by identifying factors like network-type and class imbalance that are often ignored in uniform setups, and provides recommendations for best practices based on extensive experiments with real datasets.
Link prediction (LP) is an important problem in network science and machine learning research. The state-of-the-art LP methods are usually evaluated in a uniform setup, ignoring several factors associated with the data and application specific needs. We identify a number of such factors, such as, network-type, problem-type, geodesic distance between the end nodes and its distribution over the classes, nature and applicability of LP methods, class imbalance and its impact on early retrieval, evaluation metric, etc., and present an experimental setup which allows us to evaluate LP methods in a rigorous and controlled manner. We perform extensive experiments with a variety of LP methods over real network datasets in this controlled setup, and gather valuable insights on the interactions of these factors with the performance of LP through an array of carefully designed hypotheses. Following the insights, we provide recommendations to be followed as best practice for evaluating LP methods.