Abstract: The present disclosure relates to a system and method for predicting errors in a software code base. The system generates an element set from a plurality of defect records that identify the developer who originated the defect, the code revisions related to the defect, and a classification of the defect. The system predicts the probability and type of defects in software using a machine learning algorithm configured using elements from the defect records and the code related to the defect records.

Abstract: The disclosed embodiments provide a system that facilitates user interaction. During operation, the system obtains user data for a first user and a second user connected to the first user in a social network. Next, the system uses the user data to calculate one or more influence scores between the first and second users. The system then constructs an influence graph of a set of users comprising the first and second users by creating a first node representing the first user, creating a second node representing the second user, and using the one or more influence scores as edge weights of directed edges between the first and second nodes. Finally, the system uses the influence graph to facilitate interaction among the users.

Abstract: The disclosed embodiments provide a system that facilitates user interaction. During operation, the system obtains user data for a first user and a second user connected to the first user in a social network. Next, the system uses the user data to calculate one or more influence scores between the first and second users. The system then constructs an influence graph of a set of users comprising the first and second users by creating a first node representing the first user, creating a second node representing the second user, and using the one or more influence scores as edge weights of directed edges between the first and second nodes. Finally, the system uses the influence graph to facilitate interaction among the users.