Context-aware Faceted Browsing of Semantic Web data on Android devices
(Note: this topic has already been taken)
Faceted browsing is a tried and proven way to effectively browse through large sets of data. It allows the user to look at the data from different perspectives, to better help him find what he is looking for. In other words, the user can use several data dimensions (i.e., "facets") to search for specific information. For example, imagine a user interested in Dutch artwork, who searches through the paintings in a museum. He starts by constraining the "creation location" facet to "Netherlands"; then, he constrains he "depicts" property with the value "landscape", leaving him with all paintings of landscapes created in the Netherlands. Furthermore, by constraining the "birthplace" facet of the "creator" facet value to anyone born in Italy, he selects all paintings of Dutch landscapes painted by Italian artists. This way, the user can constrain any data dimension (directly on indirectly related to paintings) in order to find what he is looking for.
Figure 1. Example of a faceted browser.
Semantic Web data is especially suitable for faceted browsing, because of the explicit semantics assigned to information. These semantics allow us to understand the meaning of information (which includes the type of the information; e.g., time, geographical data), and therefore enable us to use more intuitive visualizations for certain data. For instance, a timeline for times and dates; a map for geographical information; a tree for genealogical data or part-of information (e.g., computer parts); graphs for social networks / concept maps; etc. The first goal of this thesis is to create an Android faceted browser, suitable for browsing through large datasets on mobile devices. The student is encouraged to develop a visually attractive interface that allows for intuitive and efficient faceted browsing, utilizing a variety of facet visualizations. To get started, the student can build upon an existing faceted browsing API (created by a previous student).
Figure 2. Example of facet visualiations.
The second goal of the thesis is to make faceted browsing context-aware. This means the faceted browser should be able to correlate the browsed information with information on the user's environment. For instance, imagine the user is currently visiting a picturesque part of the city, and wants to find out whether any nice paintings of the neighborhood exist. Using information on his current location (e.g., neighborhood name, city, country), he restrains the "depicts" facet of the paintings in the dataset, resulting in all related paintings. Or, take a user looking for a nice nearby restaurant to have dinner. He can use his personal preferences (e.g., Italian food), current location, and time constraints (as provided by his personal agenda) to constrain the "cuisine", "location" and "opening hours" facets of restaurants in the dataset, respectively, in order to select all relevant restaurants. The environment and context information (e.g., location, preferences) is provided by the SCOUT framework currently under development at the WISE lab.
- Android development (optional)
- Learn about developing for the Android mobile platform
- Learn about making visually attractive, usable and effective (faceted browsing) user interfaces
- Look at current state of the art regarding context-aware faceted browsing