Overcoming cognitive and motivational barriers for networking:
contact recommendation systems in professional settings
Project description
Having the right persons in one’s network is an important precondition for receiving valuable information and referrals. Professional social network sites such as LinkedIn or Xing can help to maintain large networks. Like other social network sites, these professional networks give their users contact recommendations (“people you might know”; “people we think you might want to follow”). These cognitive interfaces aim to filter the most relevant contacts from the vast amount of potential contacts in social media networks. Current contact recommendation systems focus, however, mostly on similarity in interests or overlap in networks. This makes sense for interpersonal relationships (e.g. common friends; Facebook), but less so in professional settings, especially knowledge work, where heterogeneous networks and a higher proportion of weaker ties are more beneficial for getting access to non-redundant knowledge, creative ideas or career opportunities. Research in computer science has mainly focused on developing algorithms that extract matches by similarity. People might be happy to add former classmates, but much more reluctant to add a stranger with needed expertise. Recent psychological research has identified barriers that withhold people from instrumental networking. It is thus not enough to find the most relevant contacts; people also have to accept these suggestions.
The current project is going to tackle both problems. Computer scientists and psychologists cooperate to develop a cognitive interface that identifies heterogeneous contacts with access to relevant knowledge and presents the results in a way that reduces the cognitive and affective barriers of users. To do so, we will first conduct surveys and experiments to identify the most central barriers for accepting contact recommendations. In parallel, we will develop an algorithm that makes better (= more heterogeneous) contact recommendations. In the next step, we will experimentally examine how these recommendations should be presented so that people actually use them. The results of the project can help to connect knowledge workers in a more efficient way.
Project Team
- Prof. Dr. Sonja Utz, IWM
- Jun.-Prof. Dr. Enkelejda Kasneci, Department of Computer Science, University of Tübingen
- Lea Baumann, IWM