Logics of Representation in Structured Data Graphs Dr

Logics of Representation in Structured Data Graphs Dr. Neal Thomas | Assistant Professor, Media & Technology Studies Department of Communication | UNC – Chapel Hill ngthomas@email. unc. edu | Twitter: @lathemason

There is philosophy at work our data. Walter Pitts

Graph theory

1. The traveling salesman problem

2. Social network analysis

3. Google Page. Rank (ca. 1998)

Sir Tim Berners-Lee “The Net and the Web may both be shaped as something mathematicians call a Graph, but they are at different levels. The Net links computers, the Web links documents. Now, people are making another mental move. There is realization now, ‘It's not the documents, it is the things they are about which are important’. Obvious, really. Biologists are interested in proteins, drugs, genes. Businesspeople are interested in customers, products, sales. We are all interested in friends, family, colleagues, and acquaintances. ” 2007

Social machines today

Knowledge graphs cit yi s sh e ar m d r bo ta le ne w r e te m pe ra set or c ies ith r ur m t n e p o p nt ght of n e z citi n rre hei wi th t a l u cu or o f has auth …organize a logicist relation for datafied sociality, helping and incentivizing us to communicate in the terms of making factually correct or incorrect reference to things in the world. ion pu b lish ed by in tu re

Social graphs fri en rt e c ds on th re d iv ed ra tin f oto o h dp e g g ta rie dt d e k c e ch ar d ie l p to ce m a re we o nd e tt wi ollo is c d e unf paid …organize a performative relation for datafied sociality, incentivizing us to communicate in the terms of belonging or not belonging to a given community. me nti on ed g t a in

Predictive-analytic (machine learning) graphs Output layer …organize a signaletic or neuronal relation for datafied sociality, by reconfiguring graph data from these first two styles into a third, helping and incentivizing us to anticipate our world according to its predictions. Hidden layers ch me s re ha nti s rd bo ed Input layer f ea th er set pu bli sh ed on w or o w er auth ith n rre cu n of citize ies t a in u op p t yi n wi th cit m d e eck ion lat g paid dt ra tin ne d ie pl re to rre nt igh t le f o hoto ed p tagg o wi th rie t at ds d ar ed d en en iv ed is m c n co t er fri we cu has he re ce or ta ollo m unf by in

Knowledge graphs Philosophical outlook: Rationalism, logical empiricism, Peircean semiotics Main figures: Frege, Russell, Peirce, early Wittgenstein Semiotic mode of relation: A pragmatics of sense and reference Main affordances: Systematically moving beliefs into valid knowledge, facilitating / automating backwards and forwards-chaining reasoning • Precursors in computing: relational databases, GOFAI, knowledge engineering, expert systems • A transcendentally objectivist approach to the sign: the world as logical substance • • has he cu rre nt igh ith n m ne f ea th er set le s or ta e ar sh rd bo ies w er m rre cu n of citize w or o op tp yi n auth t ula th cit wi ion t pu bli sh ed by in

Social graphs • Philosophical outlook: Language games, phenomenology, analytic social science • Main figures: later Wittgenstein, Austin, Searle, Goffman (Schutz), White • Semiotic mode of relation: A pragmatics of illocutionary force, conversation dynamics, meaning negotiation, accountable performance in a context • Main affordances: in-group/out-group boundary management, self/other recognition in consensus • Precursors in computing: CSCW and groupware, situated action, multi-agent modeling • A transcendentally subjectivist approach to the sign: the world as social substance unf ollo nd o dc sw ed tagg pl re dt rie d ie to i ed eck ch ar a ra tin g ph o de n tte ith f oto o paid fri e iv ed d is m t er nc re ce we me nti o ne d t na

Winograd & Flores’ Coordinator (1987)

Predictive-analytic (machine learning) graphs • Philosophical outlook: British empiricism, philosophy of mind, cognitive science • Main figures: Hume, Mc. Culloch & Pitts, Rosenblatt, Hinton • Semiotic mode of relation: An asemantic empirics of impressions, sensations and associations (affects) • Main affordances: Automatic classification, decision-making. “Intelligence”. • Precursors in computing: perceptrons, connectionist AI, decision trees • Examples in everyday life: facial recognition, predictive texting, Google image search • Refuses transcendental framings in favor of one based in habit and imagination (Hume)

Guattari’s theory of the collective assemblage of enunciation

Logics of Representation in Structured Data Graphs Dr. Neal Thomas | Assistant Professor, Media & Technology Studies Department of Communication | UNC – Chapel Hill ngthomas@email. unc. edu | Twitter: @lathemason