Skip to content

Home

Jump to bottom
myeghaneh edited this page Oct 1, 2019 · 8 revisions

The aim of this project is improving the content analysis of ancient texts and books. We have used different state of the art machine learning and deep learning approach for this aim.
In this repository, you can find a refined named entity recognition (NER) model using spaCy which can annotate these labels:

  • LONG: The longitude and Coordinate in Different Formats
  • PARA: Numerical Parameters
  • ASTR:Astronomical Names
  • DATE: Date in Different Formats
  • TIME:Time in Different Format
  • STAR: Stars's Names
  • PLAN: Planet's Names
  • NAME: Names of People and Places

annotation_update_bert_04

Moreover, you can find the automatic classification of observational sentences in Kepler's book "Astronomia nova" using state of the art machine learning methods and NER. WORDCLUD

Clone this wiki locally