Liwa : News

Press release

2011-01-27T08:31:50Z

Internet pages are like stars in the sky: They are uncountable many, and every day new appear. They bring new texts, information, and pictures into existence, some of which only exist in the Internet. But who is to decide which pages are worth to preserve? Libraries and archives are currently rather helpless to deal with such gigantic amounts of data. With the current state of technology, it is not feasible to review and select all material for archiving. But this is about to change.
Two European projects will help to preserve the digital cultural heritage of the World Wide Web. The “Archive Community Memories” project focuses on automatic selection of web-content that is socially relevant. A new archiving method will not specifically search for topics or events in the Web and rate their importance. To achieve this not only Web pages of organizations or companies are evaluated, but also private content like publicly accessible blogs or social networks like Facebook. Social networks can be very helpful in discovering important Web pages, as users will suggest such pages to their friends. By harnessing such and other information, the project will help to optimize and to speed-up the reviewing process of national libraries or archives.
The total size of the EU project is eight million Euro. L3S Research Center at Leibniz University Hannover receives one million euro, and leads the scientific management. The overall management is led by researchers from University of Sheffield. There are also several other partners involved in the project, like Yahoo!, S�dwestrundfunk, and Deutsche Welle.
The project is a follow-up of “Living Web Archives”, in which researchers from L3S Research Center and other European partners have been working in the past years. The goal was to improve the quality of Web archives, especially regarding multi-media content, spam detection, as well as enabling the use of the archive for future generations.

LiWA technologies released in Open Source

2010-08-30T07:18:22Z

They are all grouped under the �liwa-technologies� project on Google code:
http://code.google.com/p/liwa-technologies/.

1� The Rich Media Capture Module - a plug-in dedicated to the capture of streaming video content:
http://code.google.com/p/liwa-technologies/source/browse/rich-media-capture
http://code.google.com/p/liwa-technologies/downloads/detail?name=rich-media-capture-plugin-1.0.jar

2� The Temporal Coherence Analyser - a plug-in dedicated to the analysis of the temporal coherence of the archived Web content:
http://code.google.com/p/liwa-technologies/source/browse/temporal-coherence

3� The Spam Assessment Interface - a Web service that enables the quality assessment of the archived Web content:
http://code.google.com/p/liwa-technologies/source/browse/assessment-interface

4� The Semantic Analizer - a component dedicated to the detection of terminology evolution:
http://code.google.com/p/liwa-technologies/source/browse/SemanticAnalyser
http://code.google.com/p/liwa-technologies/downloads/detail?name=SemanticAnalyser-1.0.zip

5� The Web Archive UI Framework - a client-side framework that helps creating User Interface helpers for Web archive browsing:
http://code.google.com/p/liwa-technologies/source/browse/web-archive-ui-framework

To learn more about each component, the Google project provides also a wiki space, giving a brief description of each module and the necessary steps for its deployment: http://code.google.com/p/liwa-technologies/w/list

You are all welcome to download and try out the LiWA components. Your feedback and comments will be greatly appreciated, helping us to improve the documentation and the usability of the technologies.

LiWA Evolution Tracking Module released

2011-04-22T12:42:31Z

The LiWA Terminology Evolution Tracking Module is a java module for Word sense evolution tracking, released under the �liwa-technologies� project on Google code:
http://code.google.com/p/liwa-technologies/downloads/detail?name=LiWAEvoTracking.zip&can=2&q=

LiWA Third Newsletter published

2011-04-07T15:36:53Z

The SHARC framework for data quality in Web archiving

2011-03-10T16:38:41Z

The download is available to download via online first in the VLDB Journal.

Web spam classification: a few features worth more

2011-03-10T16:25:40Z

In this paper we investigate how much various classes of Web spam features, some requiring very high computational effort, add to the classification accuracy. We realize that advances in machine learning, an area that has received less attention in the adversarial IR community, yields more improvement than new features and result in low cost yet accurate spam filters.

Temporal Analysis for Web Spam Detection: An Overview

2011-03-10T16:25:28Z

In this paper we give a comprehensive overview of temporal features devised for Web spam detection providing measurements for different feature sets.

Language Evolution On The Go

2011-03-10T16:24:46Z

On the Applicability of Word Sense Discrimination on 201 Years of Modern English

2011-03-10T16:22:56Z

Word sense discrimination is the first, important step towards automatic detection of language evolution within large, historic document collections. By comparing the found word senses over time, we can reveal and use important information that will improve understanding and accessibility of a digital archive. Algorithms for word sense discrimination have been developed while keeping today�s language in mind and have thus been evaluated on well selected, modern datasets. The quality of the word senses found in the discrimination step has a large impact on the detection of language evolution. Therefore, as a first step, we verify that word sense discrimination can successfully be applied to digitized historic documents and that the results correctly correspond to word senses. Because accessibility of digitized historic collections is influenced also by the quality of the optical character recognition (OCR), as a second step we investigate the effects of OCR errors on word sense discrimination results. All evaluations in this paper are performed on The Times Archive, a collection of newspaper articles from 1785 - 1985.

Talk at IWAW 2010

2011-02-01T09:14:47Z

In this presentation three use cases were presented:

* preserve Dutch public broadcasting websites (preservation of Dutch cultural heritage)
* collect Internet AV materials (mainly AV content that is broadcasted on the internet but not on traditional media)
* preserve web context (to be used by archivists for looking up relevant context information for annotating Radio & Television items)

2nd LiWA Terminology Evolution Evaluation Workshop

2010-12-29T10:31:11Z

On December 15, 2010 the 2nd LiWA Terminology Evolution Evaluation Workshop will be held in Hanover, Germany at L3S Research Center. The workshop aims at evaluating terminology evolution found inside long term archives. The workshop attendees will also evaluate the performance of the Terminology Evolution Browser, a tool developed within LiWA to better visualize evolution.

1st LiWA Terminology Evolution Evaluation Workshop

2010-12-29T10:30:32Z

The 1st LiWA Terminology Evolution Evaluation Workshop was held on March 16, 2010 at L3S Research Center, Hannover, Germany. The workshop spanned half a day and aimed at evaluating the outcome in LiWA WP5 technology.

LiWA development mentioned at FIAT 2010

2010-12-01T15:02:54Z

This poster focused on:
- an example of a shared platform Archivethe.net dedicated to heritage institutions
- archiving web video, its main issues and developments

LiWA toolkit presented to IIPC Harvesting Working Group

2010-11-12T09:17:18Z

Radu Pop presented the LiWA tools released in open-source, during the Harvesting Working Group session of the IIPC meetings.

Language Evolution On The Go

2010-11-04T11:51:46Z

Knowing about the evolution of a term can significantly decrease time needed for searching for information. It can also aid in quickly getting a broader overview, which is essential when one is on the move. In this paper we present a solution for providing language evolution knowledge “on the go”. On the 3rd International Workshop on Semantic Ambient Media Experience 2010, November 10th in conjunction with AmI-10 in Malaga, Spain, the LiWA project will present a mobile interface for easy access and visualization as well as an overview of how this evolution was found.