Richard Amiel McGough
09-19-2009, 10:16 PM
I have degrees in Mathematics and Physics, and I finished some work towards a PhD in the foundations of Quantum Physics, so I find this extremely interesting:
http://www.puttypeg.com/papers/synopticvectors.pdf
Some snippets:
2 Methods Used
The semantic vector methods used in this paper are descendants of the vector model for information retrieval, and the subsequent development of latent semantic analysis, which compresses the sparse information in the vector model’s term by document matrix into a more condensed, lower-dimensional representation. The relationship between these structures and the quantum logic of Birkhoff and von Neumann [1] has been further recognized in recent years (see particularly [2, 3]).
3 Semantic Vectors and the Synoptic Gospels
This section describes our single most deliberate experiment: testing to see if vector analysis discerns the similarity of the Synoptic Gospels. Since at least the second century AD, the Christian writings gathered in the New Testament have included four canonical accounts of the activities of Jesus of Nazareth (ca. 5BC - 30AD), and these writings, called the Gospels, have since the earliest times been attributed to authors called Matthew, Mark, Luke, and John. A basic tenet of New Testament scholarship is that Matthew, Mark and Luke are closely related, with much material drawn from one another or at least from common sources. For this reason, these three Gospels are referred to as the Synoptic (Greek, “joined eye”) Gospels.
3.1 Vector Sum Similarity
In this experiment, we set out to discover whether a semantic vector model built from the text of the King James Bible shared the view that Matthew, Mark and Luke are similar and John is the odd one out.
Methods like this may someday lead to the ability to mathematically measure the semantic correlations of all the books of the Bible to see if there are significant correlations based on the pattern of the Bible Wheel.
Richard
http://www.puttypeg.com/papers/synopticvectors.pdf
Some snippets:
2 Methods Used
The semantic vector methods used in this paper are descendants of the vector model for information retrieval, and the subsequent development of latent semantic analysis, which compresses the sparse information in the vector model’s term by document matrix into a more condensed, lower-dimensional representation. The relationship between these structures and the quantum logic of Birkhoff and von Neumann [1] has been further recognized in recent years (see particularly [2, 3]).
3 Semantic Vectors and the Synoptic Gospels
This section describes our single most deliberate experiment: testing to see if vector analysis discerns the similarity of the Synoptic Gospels. Since at least the second century AD, the Christian writings gathered in the New Testament have included four canonical accounts of the activities of Jesus of Nazareth (ca. 5BC - 30AD), and these writings, called the Gospels, have since the earliest times been attributed to authors called Matthew, Mark, Luke, and John. A basic tenet of New Testament scholarship is that Matthew, Mark and Luke are closely related, with much material drawn from one another or at least from common sources. For this reason, these three Gospels are referred to as the Synoptic (Greek, “joined eye”) Gospels.
3.1 Vector Sum Similarity
In this experiment, we set out to discover whether a semantic vector model built from the text of the King James Bible shared the view that Matthew, Mark and Luke are similar and John is the odd one out.
Methods like this may someday lead to the ability to mathematically measure the semantic correlations of all the books of the Bible to see if there are significant correlations based on the pattern of the Bible Wheel.
Richard