Abstract Search

ISEF | Projects Database | Finalist Abstract

Back to Search Results | Print PDF

Algorithm for Machine Reading

Booth Id:
CS070

Category:
Earth and Environmental Sciences

Year:
2014

Finalist Names:
Ericsson, Axel

Abstract:
The goal of this project has been to develop an algorithm to determine which adjectives pertain to which nouns in a written sentence. The algorithm is written in the PHP programming language and is optimised using various optimisation methods. I have built and optimised the entire algorithm myself with little or no help. The purpose of the project is to create an algorithm that can bridge the gap between human output and machine input by giving, in effect, the content of sentences in a parsed and machine-readable format. The finished algorithm could, for example, help translate spoken language into computer commands by outputting the content of the sentence, and the correlations between the words in the sentence, in machine readable JSON-notation. The full report and an interface I built to visualise the algorithm's output is available at: https://pack.re/nlp