• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 416
  • Last Modified:

Entity Extractor

I need to extract entities from text. I wondered if there is a entity extraction library, vcl, api, etc that can be used and incorporated into a Delphi program? I am using Delphi 2006 VCL.
0
mcmahling
Asked:
mcmahling
1 Solution
 
2266180Commented:
depends on what entities you are talking about. there are designated, specialized "extractors" named parsers like html, xml, csv, etc which will parse out the data.

for generic data you have at least one option: regular expression. but still, you will have to write the regex in order to use it (hwich i snot that hard once you get familiar with it).
0
 
mcmahlingAuthor Commented:
Thanks for your response. I would be extracting entities from .txt documents. Using regular expressions still would require me to have an ontology that must be the accessed by the regular expression to determine what words were entities. I don't readily see how to use regular expressions without that sort of semantic library. I was hoping something already existed that I would not have to do all that work.  
0
 
2266180Commented:
you canot have a computer guess what you want to extract. you need to tell it somehow. like ti or not, that's the way things work. the computer doens'tknow that a text file contains text and words and numbers and whatever. for the computer all files are the same: a serie of 0's and 1s.

not even I know what you mean by "entities". how do you want a computer to know?

regular expressions are the most egneric way of extracting arbitrary data from something. if that doesn't help, nothing does.
0
Independent Software Vendors: We Want Your Opinion

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

 
TheRealLokiSenior DeveloperCommented:
can you explain more about these entities.
What is the source text? is it just random text, or is it in an exisiting structure, such as XML, CSV?
Are theses "entities" going to be consistent within your app? Or do you really need a solution that gives you an extreme amount of flexibility?
I've written commercial EDI engines, so let me know your full requirements
0
 
mcmahlingAuthor Commented:
The entities are people, places, or things. The text is free unstructured text coming from news articles. I have heard of the open source product Annie but I wondered was their another product especially made for integration with Delphi.
0
 
Geert GruwezOracle dbaCommented:
It looks like you only need an AI to read the news articles and let the AI extract the entities ...
AI = probably human brain instead of Artificial Intelligence
0
 
MamouriCommented:
Hi

This issue related to field of Data Mining and Text Extraction. Take a look to this library: http://www.dewresearch.com/data-miner.html
The component is a general Data Mining library and could be used for a variety of Data mining usage.
0

Featured Post

Industry Leaders: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

Tackle projects and never again get stuck behind a technical roadblock.
Join Now