Automating License Rule Generation to Help Maintain Rule-based OSS License Identification Tools

Many license identification tools have been proposed to support OSS reuse. License automatically identify licenses declared in source files. Ninka is one of the most accurate tools. Because are often newly created or inherited, rules built into need be and updated on a regular basis. However, when large number unknown detected products, it not easy manually create new rules. In our previous studies, we method for clustering statements that determined unknown. this paper, propose generate from clustered statements. Our approach further filters clusters extract sequential patterns converts extracted expressions. We conducted conduct case study where applied 1, 821, 3, 561 2, 838 statement files respectively collected FreeBSD v10.3.0, Linux Kernel v4.4.6, Debian v7.8.0, confirm usefulness method. As result, confirmed successfully generated take consideration orthographical variants also efficiently identified with small Furthermore, found adding by improves licensing rule performance.