Linked Media Framework 2.2 with Apache Stanbol Integration

Last week we released version 2.2 of our Linked Media Framework (LMF). The LMF is our Open Source Linked Data Server with a rich collection of extension modules, ranging from versioning over rule-based reasoning to semantic search. The focus of this release was the integration with two important tools: (1) Google Refine as a data management application and (2) Apache Stanbol for fast entity lookup and automatic content enhancement. In this blog post, I would like to describe a bit the Apache Stanbol integration. The Stanbol integration focusses on the following major functionalities:

Data reconciliation (in combination with LMF Refine Integration): you have string data and you want to find matching resources from the Linked Data cloud for interlinking; this helps you in the task of publishing legacy data as Linked Data, as well as in enriching existing data with additional information from the Linked Data Cloud. The following screenshots demonstrates the Stanbol integration in our version of Google Refine:

Content enhancement: you have text content and you want to extract relevant concepts (e.g. locations, persons, organisations) from it; this helps you in getting more structure out of your content, e.g. for improved semantic search or data analysis capabilities. Content enhancement works either based on entities from referenced sites like DBPedia or MusicBrainz, or using the data maintained in the LMF itself (through publication in the Stanbol Entityhub). The latter functionality becomes particularly interesting when using SKOSjs (the LMF SKOS editor) to incrementally maintain a company thesaurus and use its concepts for enhancement:

Local Linked Data cache: instead of retrieving Linked Data resources from sometimes unreliable remote servers in the Linked Data Cloud, Stanbol can be used as a local cache for data from the sites it indexes; in many cases this allows a much faster and more reliable Linked Data integration, and allows using Linked Data inside company firewalls without access to the Internet.

Since the web developers building tools on top of the LMF typically do not feel comfortable with the Felix Web console offered by Stanbol, we have also implemented a configuration interface that allows simplifying common tasks like installing referenced sites, configuring enhancement chains, and synchronizing LMF content with the Stanbol Entityhub. Using Stanbol is now just a few clicks away for LMF developers. For example, the following screenshot shows how preconfigured enhancement chains can be installed in Stanbol:

A very nice scenario is using LMF + Stanbol for Semantic Search. The LMF includes a very powerful semantic search component. With the Stanbol integration comes a new option to send the texts that are indexed to Stanbol to extract entities and use this additional information to extend the search index with information from the Linked Data Cloud or the local thesaurus. The LMF semantic search component also supports incremental configuration of both the search index and the user interface. This allows for very fast “experimentation cycles”, where you improve both your search configuration (e.g. adding additional Stanbol-based enhancements) and the user experience very quickly. The following screenshot gives a preview of our search interface editor:

Download today! The LMF including Google Refine and Apache Stanbol is available for download in a bundled installer that you can use for installing the whole package on your own computer. Go to the LMF Downloads section and try it out!



Author: Schaffert

Sebastian Schaffert is currently head of the knowledge and media technologies group (KMT) at Salzburg Research with about 18 members, and is working there as senior researcher and project manager as of August 2005. Sebastian Schaffert is engaged in research on the Social Media, the Semantic Web, Knowledge Representation, and Multimedia Information Systems.

Comments are closed.