Extracting signal from the noise of app reviews

Apps stores offer a transparent feedback loop between users and developers via public app reviews and direct distribution of updates to users. Significant effort is expended to glean useful information from reviews and to examine this feedback loop. This thesis uses statistical and text analysis on 8.7 million reviews to demonstrate that the majority of reviews, when read in isolation, do not yield developers much actionable feedback. Reviews are analysed to offer app benchmarking, search space reduction techniques and typical review properties to detect and sort useful reviews, toward informing project expectations, priorities and scoping.