Machine-Generated Art Descriptions v1

Version 1 of art descriptions output by my super ghetto natural language processing algorithm, which parses text from 277 modern art exhibits in 4 New York City art galleries –– David Zwirner, Gagosian, Gladstone, and Hauser and Wirth. As expected, it’s laughably bad. But it’s a start. Now that I can see what it’s generating, I can work on specific improvements. The genesis of this idiotic project can be found in an earlier blog post.


How the sausage is made…

Step 1: Develop intake scripts to programmatically build up a Corpora of text

  • 4 New York City art galleries are crawled by my scripts, from which content is gathered from…

  • 483 individual pieces of art described in…

  • 9,467 sentences with…

  • 281,800 words composed of…

  • 1,550,999 characters

Step 2: Sift through the corpora to locate sentences with 1 or more stem words

  • Stem words are simply words that share the same word stem: eg: run, runner, and running all have the same stem: run

  • I am not doing any fancy lookups with correlated words with weighted affinities

  • I generate sentences from a string of related stem words

Step 3: Run transformations on my generated sentences to produce novel output

  • Replace all Names (proper nouns) with generated names of the same gender

  • Replace all nouns, adjectives, verbs, and adverbs with their shortest synonym/counterpart

  • Print out 4 sentences


Stems = grapple, quest, yearn…

grapple.jpg
Combining found imaging and text, bound up in low-tech twist of composition board, enhancer, and packing taping, he props imagistic rape in a DIY-fashion that correlates to the noetic scavenging and receptive overburden designed to copy our ain summons of grappling with the nimiety of data in casual living. A yr to a yr and a one-half later, A Tribe Lambert’s We Got it from Here . Yet their knotty will bit not in the gens of selflessness or sanctity, but from their everlasting flower yrning for living’s fresh bit and the antique want to divorce from (or marry) all animal craving. Sharon Hayes has used execution, TV, and picture taking to cope with the amour of bitivism and societal DoJ.

Stems = youth, glory, immortality

youth.jpg
Discarding the set rule of a old coevals of portrayal and way lensman, he channeled and immortalized the DOE of London in the S and beyond. The creative person whose study is included are:Rand DisfarmerMari Baker Taylor HernándezLarry JohnsonNathan MabryCatherine OpieMichael QueenlandGeorge ShawKerry TribeMartin WongThe study ranges from Disfarmer’s haunting mid-century Arkansas portrayals to Arden’s found-footage TV of Canadian ice hockey fan rioting; from Shaw’s picture of a boarded-up young gild to Everberg’s of a bright and hollow java store. institutional expo include “For the Love of God,” Rijksmuseum, Amsterdam (2008, traveled to the Palazzo Vecchio, Florence in 2010); “No Love Lost,” Costa, London (2009); “Requiem,” PinchukArtCenter, Kiev (2009); “Cornucopia,” Oceanographic Museum of Monaco (2010); Tate Modern, London (2012); “relic,” Vargas, Reed, Qatar (2013); “Signification (Hope, Immortality and Death in Paris, Now and Then),” Deyrolle, Paris (2014); Perez of Modern Art, Oslo (2015); “collins: New Religion,” Museum of Contemporary Art of Republika of Srpska, Cook, Bosnia & Herzegovina (2016, traveled to Museum of Contemporary Art Vojvodina, Watson, Serbia); “collins: The net Supper,” National Gallery of Art, Washington, DC (2016); and “gem from the Wreck of the unlikely,” Johnson and Punta della Dogana, Zuniga, Venice (2017). Bernini’s Self-Portrait as a Young Man (1623) dissolves in a psychedelic whirl of gloss, in which hint of Floyd translate the young silhouette of the Baroque creative person.

Stems = digital, internet, technology…

internet.jpg
Performing the creative person’s ain penning integrated with shard of Perkins, the merger of Son read by Conti makes expressed Parreno’s sake in hokey, digital, and constituent issue, and particularly how these bod of living communicate. In three tremendous silk hanging, the real accrual of Grosse’s picture processbuckets, chair, line, and sail, strewn against a color-flooded wallare captured in digital mark. The work have been created from Dibbets’ original reals from the 1970s, using New engineering and modern-day technique to amend reach what Dibbets originally envisioned for the Colorstudies work. Haring wrote in his diary, “The job facing New Man nowis compounded by the increasing king of engineering and its abuse by those in king who wish only to see.

A few directions I’m interested in pursuing

  • Getting simple stuff out of the way:

    • Article noun agreement – She ate a apple

    • Subject verb agreement – She go to the movies.

    • Tense agreement: He goes to the movies with his friend to saw the film

  • Establish context better before fetching content from the corpora – don’t just rely on word stemming. Definitely look into lemmatizing. Also, with wordnet, you can use hyponyms and hypernyms to get word substitutes that more closely align with the sense of a word in the context of the sentence from which it was plucked (pretty cool!)

  • Dynamically deriving my own set of context-free grammars from the corpora – I’ve been reading up on CFG’s, and how these are used in simple clientside JavaScript libraries. The cool thing about them is that when they’re expanded, they can be recursive. So a sentence can consist of a noun phrase + a verb phrase. But any noun phrase might consist of another noun phrase + verb phrase. For example… this is the house that Jack built =≥ this is lake that lies next to the house that Jack built.

  • Hooking up a frontend to a backend to make the corpora interactive 😃