Haiku Generator that Reads Blogs and Illustrates Them with Sounds and Images

Since paintings produced by the Aaron system [Cohen, 1995] were shown at the San Francisco Museum of Modern Art in 1979, artificial art has become a widely discussed topic that is In this paper we introduce our haiku generator, not limited to computer enthusiasts. However, the main question which, in contrast to other systems, is not restricted remains unanswered - was Aaron, the plotter-equipped to limited classic vocabulary sets and preserves art generating program, extending its author's "teachings" a classic style without becoming too random and and actually creating something, or was it a mere "expert abstract because it performs a semantic integrity system" with newer capabilities incrementally added to its check using the Internet. Moreover, it is able to repertoire? The problem, at least in our opinion, lies in subjective analyze blog entry input and, by using nouns and evaluation, which until recently was widely avoided adjectives for web-mining, to stay on topic and still in science and engineering. However, we think that the closer preserve kigo, traditional seasonal words used in that machines are to humans with their endeavors, the less Japanese poetry. The haiku generator utilizes grammar uncanny uneasiness we will feel with machines that paint templates automatically generated from poems [Lindemeier et al., 2013], sing [Fukayama et al., 2010], joke written by Japanese poets and a lexicon of 2,473 [Dybala et al., 2008], generate stories [y Pérez and Sharples, kigo words from an online haiku repository. In 2004] or write poetry [Colton et al., 2012]. We believe that, addition to generating haiku poems, it can output similarly to how the discussions about the meaning of "life" them vocally together with related sound effects after the discovery of DNA calmed down, the meaning (or and images retrieved from the WWW. Our experiments rather the breadth of interpretation) of words like "creativity" demonstrate that the proposed system generates will also evolve, and one day we will agree that machines can high-quality haikus and that using contentrelated produce art that we enjoy and do not recognize as "artificial".