TUD Logo

TUD Home » ... » Research Projects » Finished Reference Projects » Finished Reference Project

Computer Networks

Finished Reference Project

Project:

Fedseeko - Federated Semantic Product Information Search on the Web

  • Internet Information Retrieval
  • Product information
  • Ontology Matching
  • Web Scraping

Overview:

In the quest of finding product information, the WWW nowadays offers a broad and up-to-date information set that is freely available for every consumer. Product information is put into the net by different online malls, the producers themselves, as well as a big and heterogeneous group of third parties, including blogs, boards and testing websites. Unfortunately, the enormous amount of information also holds some drawbacks. On the one hand, it is a quite complex task to localize the desired information. On the other hand, the retrieved information might be wrong, out-of-date or biased, e.g. concerning advertising. Fedseeko is a Ruby-on-Rails-based Web application offering the consumer a central point of truth for retrieving product information. Fedseeko aims at creating an all-embracing view on the product of desire using product information from all three mentioned information source types. Different technologies being located in the backend like Web services, information extraction (especially Web scraping), natural language processing, ontology matching or ontology learning are used for retrieving and preprocessing the information. The frontend is designed using technologies like AJAX, Javascript frameworks and RHTML. The application is based on JRuby and thus offers the possibility to include Java classes. Eventually, browser plugins are employed for displacing extraction tasks on the clients’ side.

» Projects homepage


Project leader:

Dipl.-Medien-Inf. Maximilian Walther


Project staff:

Dr.-Ing. Daniel Schuster
Dipl.-Medien-Inf. Maximilian Walther


Resources:


Last modified: 9th Dec 2018, 11.29 PM
Author: Webmaster