Data Extraction (work in progress)

This page will contain demonstrations of a couple of tools we've been developing.
Comments and suggestions are most welcome - please contact Dr. Yoram Cohen.

Please note:

  1. The examples were only tested in Internet Explorer and in Firefox.
  2. The map application will only work in Internet Explorer and Firefox version 3.5 or above.. In Internet Explorer, after selecting the map link, please click "Allow blocked content" on the upper part of the screen.

Naming Patterns

Motivation: Naming patterns and name distribution from one generation to the next are important criteria for determining, among others, preservation of family- and self-identity and religious traditions. In the present case, our tool extracts data relevant to the identification of father-son naming patterns.

Implementation: This tool is based on a python script that locates instances of phrases such as "X son of Y" in the texts, and keeps note of the two names along with their family relationship.
The script's output is parsed by a Java Servlet, which extends the DataSourceServlet class of the Google Visualizations API. The client runs a javascript which contacts the server to retrieve this data, parses it and displays it to the user.
Please note that the script does not yet read the lemmatization data of the ATFs to decide on the ethnicity of each PN. This ethnicity is currently inserted by hand for the demonstration.

Comments: We hope that the extraction of names and family relationships will be replaced in the near future with a utility common to all the Oracc-based projects.

Demonstration of the tool

Geographical Distribution

Motivation: By placing the individuals on the map through the aid of Google Maps, a sharper picture of their settlement through space and time can be achieved. The data can point us to the settlement distribution of the exiles all across the Ancient Near East at the times of the Assyrian, Neo-Babylonian and Persian Empires.

Implementation: This tool retrieves an XML file from a Java servlet. The XML encodes the location of the different toponyms along with the individuals who are related to each toponym. Currently, this XML file is not automatically generated from the texts, but is rather manually created. A client-side javascript parses this XML, and uses the Google Maps API to display the information on a map.

Demonstration of the tool