Aspect mining algorithm development is a field that is slowly starting to gain more interest in the general Aspect Oriented Programming (AOP) community. However, it is a field that suffers from many limitations stemming from the fact that the definition of an aspect is very abstract. Due to this, the researchers working in the field have not had a solid algorithm development or evaluation methodology. This has led to a lack of collection direction, and it has made it exceedingly difficult for new researchers entering the field to decide how to proceed. The lack of a concrete evaluation methodology makes it impossible to judge what are promising research directions to pursue for developing aspect mining algorithms.In this thesis, we have proposed a concrete algorithm development and evaluation methodology based on the context that aspect mining algorithms will be used in. We have proposed a new evaluation metric, the Relative Uncertainty metric, which allows us to compare different algorithms developed for aspect mining. We have developed an aspect mining tool, Quantum, which can be used as a development platform for creating new aspect mining algorithms. We have used Quantum for the development of 8 aspect mining algorithms, and we have presented a detailed evaluation of these algorithms. The results illuminate promising directions for the development of aspect mining algorithms.