View source for BDP: Big-Data Poetry

Big data is a buzzword, as is cloud computing. But the data science and network-clusters behind both of these terms present extraordinary viable unprecedented computationally-tractable opportunities for language processing and radical poetry generation. In the summer of 2014 I took an intensive 11-week course in data science programming using Python. Based upon this theoretical and practical coding knowledge, I produced http://bdp.glia.ca, research where I apply a combination of data visualization, language analytics, classification algorithms, entity recognition and part-of-speech replacement techniques to a corpus of 10,557 poems from the Poetry Foundation, 57,000+ hip-hop rap songs from Ohhla.com, and over 7,000 pop lyrics. Currently the poems generated lack thematic structure. I propose to read extracts and reveal intricacies from http://bdp.glia.ca/ Including rapi improv-free-styling from the real-time output of the system (SPREEDE https://vimeo.com/105819691) Roland Barthes famously predicted the death of the author: yet I do not think he foresaw the cause of death as big data. And I doubt Barthes intended to imply the irony that from every death there springs new life. It seems plausible now to suggest that writers are sets and repertoires of techniques and preoccupations. And each writer writes within a cultural context, a time, a vocabulary, and a tradition. Once these traditions are mapped, propensities or paths for future writing will be either generated or grown as variations to assist authors in exploring creativity that conforms to their innate self while at the same time assisting them to see opportunities. The author will not die but expand to explore more of their potential using a computational symbiont. Data science algorithms are capable of finding topological patterns within languages, and thus poetry. By examining which patterns fit cores of genres and which are outliers, notions of creativity and modes of writing will necessarily shift: exploratory writing will swiftly outgrow the uncreative mode of pure appropriation and move toward nuanced expressive augmentation of the writer’s own persona. Big data, I claim, has equivalent power not just to depersonalise but to repersonalise. 

(Source: ELO 2015 Catalog)