Using change point analysis to determine perception accuracyinsocial media opinions
Sallis, P; Claster, W
Permanent link
http://hdl.handle.net/10292/8973Metadata
Show full metadataAbstract
Social media such as Facebook and Twitter have become repositories for extremely large volumes
of data representing for the most part, personal opinions and statements of individual users of these
instruments for interpersonal communication. They provide a forum for discourse and debate. The strings
of words broadcast by individuals are predominantly casual expressions reflecting personal attitudes or
sentiments. Due to the pervasive nature of social media and the opinions they proliferate, it is assumed
likely that these casual expressions are being relied on by some readers of the discourse, for their decision
making influence. This paper describes how Change Point Analysis (CPA) can be used to bring some
precision to the plethora of casually expressed sentiment sourced from social media databases. It is
contended that where the conjunction of opinions expressed are related to a factual occurrence in time and
space, a change point can be established, which indicates the potential reliability on a given sentiment as
being a reliable influence factor for opinion formulation or other decision. When related to tourist
destination data, the case study for Japan described here is convincing in its analytical result exposition