Institute for Web Science and Technologies · Universität Koblenz - Landau
Institute WeST


Example applications enabled by WeST knowlege.

Text Mining

One use case for text mining at WeST is an urban maintenance tool for citizens that allows them to report issues to their municipality. A high percentage of all issues have not been assigned to a proper category by the citizen reporting it but are assigned to the default category “others”. In the past, uncategorized issues lead to difficulties and additional workload in the further procession steps by the city council. Therefore, a proper automatic categorization with meaningful categories is of great interest.
Our current work consists of combining sources of background knowledge, like Wikipedia, and state of the art techniques to gain new insights in the  issue data set. As an outcome, we developed an extended version of the ESA algorithm and applied it to the issue reports. There, we could already reduce the number of uncategorized issues and thus the workload of the employees in the city council.

Eye Tracking

Eye tracking data can be used in the generation of information about photos. The gained information could hardly be generated by means of fully automatic algorithms, such as which parts of the photos show specific objects (even for objects with unusual visual features) or which photos of a collection are interesting to specific users. The aim is to exploit natural human viewing behavior as it is available without additional effort and provides valuable information.

Politics - Analysing Liquid Feedback

Liquid Feedback is a voting system for collaborative decision making. Users can create and vote for proposals and discuss them. Additionally, votes can be delegated to other users. The Liquid Feedback dataset combines text documents (proposals), discussion boards, a delegation network, a voting record and can additionally be linked to twitter and the wiki of the German pirate party via user names. Therefore techniques of network analysis (e.g. (un-)link prediction), data mining, natural language processing etc. can be employed and combined for analysing the data.

Semantic Web - eLISA: enhanced Local information, search and aggregation

The attractiveness of a city is a very subjective assessment that bases on different criteria depending on the age and life circumstances. Often it is difficult to bring own ideas in accordance with local conditions. Also, there are many different factors that can play a role in assessing a place. eLISA allows the user to select individual important indicators, prioritize and store them in a user profile. Then an individual location-based attractiveness factor is calculated and clearly displayed on a map. The user can see at a glance the streets, neighborhoods or regions that are interesting to him because they are marked accordingly. This allows arbitrary regions to be analyzed quickly and easily using various individual criteria.

Semantic Browsing of Multimedia - SemaPlorer

SemaPlorer is an easy to use application that allows end users to interactively explore and visualize a very large, mixed-quality and semantically heterogeneous distributed semantic data set in real-time. Its purpose is to acquaint oneself about a city, touristic area, or other area of interest. By visualizing the data using a map, media, and different context views, we clearly go beyond simple storage and retrieval of large numbers of triples.