INTERTAIN Experimental Research Lab is an initiative by the Department of Computer Science at the VU University Amsterdam, the Netherlands. The Home Lab provides an environment for key research themes which involve user-centered, distributed, interactive and intelligent multi-device applications. See the about page for more information. Below you can read about events and studies in the lab.
eMate: building an iPhone app for lifestyle support
Masterproject / minimasterproject / paid project
It is no big secret that adhering to a healthy lifestyle can be quite difficult. It requires one to eat healthy, exercise regularly, and, in some cases, even take medication on a regular basis. This can be quite difficult: people find it hard to stay motivated in all circumstances. As professional personal coaching is expensive, it seems that automated intelligent support might offer a solution. Smartphone apps can offer dynamic and personalized support when it is needed the most. They can communicate with the patient and can offer motivation or help.
In this project the student will develop an iPhone app that is able to do exactly that: helping people to adhere to their healthy lifestyles or therapies. The app will be modeled on a working Android version of the app, which already incorporates the key features necessary. These key features include:
- asking the user questions about daily goals of exercise, diet, and medicine intake
- sending reminders to take medication
- sending motivational messages that encourage the user to stick with his/her goals
- providing the user with status updates
The app will connect to a database that provides the texts for the messages and contains information about the user.
We are looking for someone with hands-on experience with:
- programming (preferably in Objective C as this is what is used for iPhone programming, but not a strict requirement)
- MAC OS X
- SQLite
- JSON
Interested? Or would you like to know more about this project? Please send an e-mail to a.van.wissen@vu.nl !
Social Web Software
For performing experiments on Social Web Data there are a number of software tools, which can be used. Typically, setup can be a long process and not always straightforward. We need to have a package created of selected tools and datasets, with clear installation procedure included (for students to use on their own computers). In the computers in the lab the pre-installation of this software package is also needed.
Expected effort about 3 months
Contact: Lora Aroyo l.m.aroyo@vu.nl or Marieke van Erp marieke@cs.vu.nl
WAISDA? Video Tagging Game
The Waisda? (which translates to What’s that?) video labeling game was launched in May 2009. It invites users to tag what they see and hear and receive points for a tag if it matches a tag that their opponent has entered. Waisda? is the world’s first operational video labelling game. The underlying assumption is that tags are most probably valid if there’s mutual agreement. Waisda? managed to attract a large audience since its launch on the 19th of May 2009 the game was consulted by 9.198 unique visitors, gathered 340.551 tags describing 604 items, and the tags were added by a total of 2.296 players.
The first pilot of the game in 2009 was running with videos from the series Boer Zoekt Vrouw. The current pilot will run with Man Bijt Hond. You can be involved in various activities:
- setting up the game at the Intertain Lab computers
- working with the current code to set up the old and the new interface with the same videos
- creating bots to play automatically the game
- working on user motivation strategies to play the game longer and more often
- working on gaming strategies to increase the quality of the collected tags
- developing ideas for improving the gaming interface
- developing the post-gaming interface, where users can score additional points to improve their tags
- working on the analysis of the collected tags and their mapping to different vocabularies
- working on maintaining the community of taggers and players
- any other interesting idea you have in mind in the context of this project
Check out our current demo: http://waisda.nl/ (the more people log in at the same time, the more fun it is)
Duration: depending on which topic is chosen, average work is about 2-4 months
Contact: Lora Aroyo l.m.aroyo@vu.nl or Michiel Hildebrand michielh@few.vu.nl
Internet Barometers: Online Activist Networks
The tactical repertoires of activist groups are broadening and the Internet offers an instant platform for debate over firm – activist group relationships. We will use data available on the Internet to study the way activist group networks try to target companies on issues of corporate social responsibility (CSR) on the
web. We focus on activist networks around SOMO, the Centre for Research on Multinational Corporations.
The project aims at thee related activities:
- Data collection and data modeling to compile hyperlink networks, generating semantic maps and tracing implicit frames.
- Developing Methods for the analysis of activist groups’ tactics (a tactics database) and the networks they form. In doing so we want to combine hyperlink networks and semantic map analyses but also aim to get beyond these methods.
- Data analysis and interpretation to visualize large amount of data to end users for personal and group analysis
We are especially interested in involving students with interests in some of those (should not be all at the same time:
- social network analysis
- information visualization
- building APIs
- user interfaces
- hacking
- crawling data from various social network sites
- any other interesting idea you come up with in this context
Contact: Lora Aroyo l.m.aroyo@vu.nl
Places of Interest: Personalized City Guides
Amsterdams Museum (AM, former Amsterdam History Museum) is leading a project called Places of Interest, which should serve as an online Michelin Guide for interesting cultural, historic and touristic places to visit, discover and visit around Amsterdam. The project is developing an online platform where places (locations) of cultural and historical significance are central. We will link objects from different cultural heritage collections, e.g. Amsterdam Archive, National Monuments, Beeld & Geluid (Sound & Vision), etc. to actual physical locations in Amsterdam. An important part of the project is to provide the information in multiple languages and adaptable to various tourists needs.
Current aim is to develop:
- a mobile guide with a route planner with cultural and historical information – a location-based application for smartphones
- a multi-touch table application as a board game to explore cultural sites in Amsterdam
- a mobile tablet application combining aspects of the game and the mobile guide
You could work on any of the above tasks, including:
- developing innovative tour generation strategies for the mobile guide
- developing interactive tour guide or gaming interfaces for different platforms, e.g. mobile, tablet, touch table, interactive screens, PCs
- working on APIs for the data access, search and presentation
- any other interesting idea you come up with in this context
Duration: depending on which topic is chosen, average work is about 2-4 months
Contact: Lora Aroyo l.m.aroyo@vu.nl or Victor de Boer v.de.boer@vu.nl
Related project: AGORA
- Check out these videos for Amsterdam DNA: http://www.amsterdammuseum.nl/tours
- Check out the blog of the project: blog.plaatsenvanbetekenis.nl
- Watch a video, which illustrates some of the inspiration for this project: http://www.youtube.com/watch?v=c-FgmZZanSc
AGORA: Hacking History
The Agora project is a collaboration between the History and Computer Science departments at the VU, the Rijksmuseum Amsterdam and the Dutch national audiovisual archive Beeld en Geluid. The aim of Agora is to develop a social platform in which museum objects can be placed into an explicit (art)historic context & to allow users to create their own personal narratives in this digitally mediated public history. We model historical events, so that they can be used for linking objects from different museum and archive collections, e.g.
- show me all the artefacts related to the second world war;
- show me how they are related, etc.
In this project you could work on:
- development of innovative interactive search & browsing interfaces
- development of a social platform for exploring history narratives and matching objects with people
- development of games or quizes using the data generated in the project
- improving named entity and event extraction with background knowledge
- development of historic event thesaurus/ontology
- development on crawlers of historic events around the Web
- any other interesting idea you come up with in this context
Related projects: Places of Interest.
Expected effort: depending on which topic is chosen, the average work is about 2-4 months
Contact: Lora Aroyo l.m.aroyo@vu.nl or Marieke van Erp marieke@cs.vu.nl
- Check out our demos: http://agora.cs.vu.nl/demo/
- Watch this inspiration video: http://www.youtube.com/watch?v=CoDPC9YU5lM
We appreciate students with general ‘hacker’ attitude, experimental & exploration spirit, as well as curiosity & pragmatic development attitude.
COMBINE Project
The work that needs to be done is design, programming and experimentation with a new toolkit, let’s call it the COMBINE [1], that will do selective crawling for sense making tasks. That is a special kind of information retrieval task where you don’t know what you’re looking for. Something has happened and you have no clue to what happened, but step by step, aided by the system that in the background selectively gathers more and more knowledge, you make a sketch of the situation and find out what it was that you wanted to know.
In a sense, it is not Question Answering, but Answer Questioning, because it helps you to come up with the right questions. As opposed to relevance feedback it does not change the query to better suit your information need, but it (also) changes your information need to lead you to the most salient information.
We will try out the COMBINE to make sense of disruptive events, like environmental and natural disasters, riots, and armed conflicts like maritime piracy. Given very minimal descriptions we will try to guide the user to the complete storyline of what happened.
Tasks include, but are not limited to:
- Working on an interactive information retrieval user interface.
- Working on a context sensitive instance matching engine that can match events, actors, places, etc.
- Working on improving named entity and time extraction with context information.
- Working on a selective distributed crawler suite that makes use of existing search engine APIs to seed the crawler.
contact: Willem Robert van Hage W.R.van.Hage@vu.nl
NoTube: Interactive Table
We have interactive touch-table available in the lab. The table runs on a Windows machine with a software for calibration and multi-touch interpretation. There are several projects at the VU which consider multi-touch table as a possible interaction interface with users. Within the NoTube project (dealing with converging Web & TV) there is an opportunity to explore a “Multi-touch Javascript/HTML5 video UI. Several points of exploration are:
- to work in a fixed browser (e.g. Firefox 4.x, Chrome etc.)
- to use Javascript XMPP API (e.g. Strophe.js toolkit)
- to communicate to the ‘big TV/projector screen’ for display
- to play videos ‘locally’ on the table as a preview, or show thumbnails
- to realize a multi-touch integration with table using either http://pooky.sourceforge.net/wiki/PookyTouch (a Firefox addon) or new multi-touch built-in support from Firefox
- to build on rough prototypes we have already in NoTube (e.g. we have single-touch thumbnail navigator), and rough XMPP message protocol
The focus of this project will be on exploring the multi-touch table environment as it is right now, and updating it to the current evolvement of the available technologies. Additionally some hacking needed for the ‘remote controllable’ device (although we have some code / demos).
Possible skills needed:
- familiarity with Javascript, HTML and browser-based Web application development.
- general ‘hacker’ attitude, experimental pragmatic approach
Expected effort: depending on specific implementation, around 2 monthts.
Contact: Lora Aroyo l.m.aroyo@vu.nl or Dan Brickley danbri@danbri.org
- Watch a video, which illustrates some of the inspiration for using multi-touch table: http://www.youtube.com/watch?v=c-FgmZZanSc
- And this older one too (on which we worked some time ago): http://www.youtube.com/watch?v=tsLjGyZ_RCs
Interactive Table software/hardware
We have interactive touch-table available in the lab. The table runs on a Windows machine with a software for calibration and multi-touch interpretation. Currently there are two example applications for this table – (1) is a standalone application in Flex and (2) a web-based application using Pooky add-on for Firefox. These are only two examples of how to build applications for a multi-touch table. We are looking for enthusiast students who are interested in exploring what are currently different options to develop interactive applications, e.g. programming, libraries, formats, etc. The project will require an exploration of current technologies and will also produce a setup installation (plus) documentation for a selected solution for this table.
Related Interactive Table projects: NoTube Interactive Table. You can also think of interesting interactive touch table demos in the context of the AGORA and Places of Interest projects.
Expected effort: depending on specific implementation, around 2 monthts.
Nabaztag
![]()
In the INTERTAIN lab we have 10 Nabaztags available – a Wi-Fi enabled ambient electronic device in the shape of a rabbit. Unfortunately, the company that produced them went bankrupted and the rabbits cannot do anything anymore. Luckily, there is some initiatives in the open source community to develop an alternative implementation which would revive the animals.
The goal for this project is to investigate the status of alternative software that can be used with the Nabaztags, and – if feasible – develop a demo application for the Nabaztags, preferably usable for high-school activities.
The expected amount of time for this project is 40 hours for the initial investigation and 120 hours for development of a demonstrator.
