1. Salzburg Research IOT Avatar Salzburg Research IOT
  2. Untitled project
  3. lmf

Wiki

Clone wiki

lmf / Google-Refine-Users-Documentation

View History

Documentation for user of the Google Refine LMF Extension

[[Google Refine Extension|Google Refine LMF extension]] documentation for users

Features

General documentation

Installation

The LMF installer provides a version of refine with all the necessary already installed. But if you'd need/prefer to do it by yourself, these would the necessary steps to follow:

In upcoming releases, the full tools will comes with the LMF installer. Meanwhile you'd need to manually install it.

  1. So first you'd need to install Google Refine
  2. Then install the RDF extension.

Once you've completed these prior steps, you have to proceed in the same way for installing the LMF extension: 1. Go to the extensions directory under the Google Refine workspace. 1. Download the LMF extension (@@TODO@@: provide extension bundle) 1. Unpackage the downloaded file to the "extensions" directory you already located at the first step 1. Restart Google Refine, and you should be able to see the LMF menu at the extensions' toolbar.

http://wiki.lmf.googlecode.com/hg/images/refine/lmf-menu.png

Usage

After install the LMF Refine Extension, you could see a simple customization of the style, besides some new features available under the LMF menu at the extensions' toolbar.

http://wiki.lmf.googlecode.com/hg/images/refine/refine-lmf.png

http://wiki.lmf.googlecode.com/hg/images/refine/refine-lmf-project.png

= LMF Configuration =

The first thing you should need to do if the configuration of your LMF instance: under which URI is running and authentications data if required.

http://wiki.lmf.googlecode.com/hg/images/refine/lmf-configuration.png

= RDF mapping =

Then you would need to edit your RDF skeleton.

http://wiki.lmf.googlecode.com/hg/images/refine/edit-rdf-skeleton.png

= Add reconciliation services based on Apache Stanbol =

The Apache Stanbol EntityHub provides efficient access to entities' indexes. Since the latest version implements the Reconciliation Service API, it could be used in Google Refine for an effective reconciliation. Read more about this integration.

http://wiki.lmf.googlecode.com/hg/images/refine/add-reconciliation-stanbol.png

http://wiki.lmf.googlecode.com/hg/images/refine/stanbol-registration.png

http://wiki.lmf.googlecode.com/hg/images/refine/stanbol-registered.png

http://wiki.lmf.googlecode.com/hg/images/refine/stanbol-reconciliation.png

In this explanation we used the IKS demo server which could be not available or cope with multiple people doing reconciliations. Therefore you should provide your own instance of Stanbol running the necessary indexes for your case. For instance, adding additional indexes, such as DBPedia, you would be able to quickly perform reconciliation against the indexed entities.

= Save to LMF =

you don't need to manually export the RDF data to a file; the extension allows you to directly save the data into the LMF instance previously configured, placing it available as Linked Data without any additional effort.

http://wiki.lmf.googlecode.com/hg/images/refine/save-to-lmf.png

Updated