Wandora is a tool for people who collect and process information, especially networked knowledge and knowledge about WWW resources. Our goal is to provide an easy way to aggregate and combine information from various different sources and to allow the user to play with and manipulate the collected knowledge flexible and efficiently, and without programming skills. Wandora is not only a graph database but a collection of easy-to-use tools to collect, manipulate and publish the information.
Wandora is a desktop application, written in Java, and runs in Windows, Linux and Mac OS X computers. Wandora has graphical user interface and offers several topic map visualization models. You can view the topic map using a topic map browser, graph and treemap visualizations, or you can program your own visualization with embedded Processing language. Even more visualizations is available with the embedded HTML browser. Several visualization windows can be viewed simultaneusly. Next image views Wandora with three topic panels. Middle visualization is created using a Processing script, upper-right is Wandora's default graph visualization and lower-right is default topic browser.
Wandora's internal data model is layered. Layered knowledge base builds upon separate information layers each containing only part of the knowledge. You can find analogous layering paradigm in most modern graphics editing programs where the user composes the image with transparent layers each containing only small part of the complete image. In Wandora the user can insert topics and associations in separate layers and view only layers that she finds interesting. It is easy to manage separate information layers. Wandora supports not only memory based information layers but also database based layers.
Wandora's internal database is based on Topic Maps and Wandora can read topic map files in XTM2, XTM1, LTM and JTM formats. Wandora reads also RDF data in XML, N3 and Turtle formats. Wandora shares many properties of semantic web and linked data tools. Moreover, Wandora features an OBO flat file import. Next image views Wandora's import options at the moment (2013-05-22).
Addition to obligatory topic map exports, Wandora can export information in various widely known graph formats: Graph Modeling Language, GraphML, and GraphXML. Some exports are available as HTTP services provided by Wandora's internal WWW server. The user can export not only complete information layers but any topic selection also. Wandora's internal WWW server provides not only data exports but also several D3 based visualization. Experienced user can also build easily her own service modules for the internal WWW server. The developer can also export a topic map as a collection of static HTML pages. Next image views a Firefox browser with a web page produced by Wandora's internal WWW server.
What makes Wandora an unique and important information tool, is an extensive collection of information extractors. These little tools can be used to transform files into topics, distill entities out of text, link entity to it's superclasses and find synonyms for a term, for example. While these extractors all generate topic map stuctures, the developer can truly sculpt the knowledge. For example, Wandora features Alchemy API extractors used to entity, keyword, and sentiment extraction; Twitter extractor used to convert Twitter messages and related to a topic map graph; and several Excel extractors that convert spreadsheets data into topic maps. Wandora Team has also created a Firefox and Thunderbird extension that can be used to perform extraction directly from WWW browser and email client. Next image views a dialog of Wandora's SPARQL extractor. It is used to transform SPARQL result sets into topic maps.
Wandora contains a huge collection of topic map editing features. You can not only edit topics and associations directly but also transform topic's internal structures -- such as occurrences to other structures -- such as variant names. You can also choose to apply editing feature to not just one topic but a collection of topics which speeds up topic map refactoring.
Wandora compares topics maps and features a topic map patcher that can be used to generate topic map transformation scripts. Wandora calculates topic map diameter, connection count and distribution and clustering coefficient. Wandora supports both Wandora specific Query Language and TMQL. R language integration makes Wandora a data source for R environment. Wandora has also several topic map generators that create topic map graphs algorithmically. For example, Random graph generator, Tiling graph generator and L-system generator are such generators.
To conclude, Wandora is an open source, developer friendly and powerful information gathering and manipulation tool that has an extensive collection of useful features. Now you might want to continue to