COMPUTATIONAL METHODS FOR INTELLIGENT MATCHMAKING FOR KNOWLEDGE WORK – CASE CMAD Jayesh Prakash Gupta, Tampere University of Technology, Finland Jari Jussila, Tampere University of Technology, Finland Ekaterina Olshannikova, Tampere University of Technology, Finland Karan Menon, Tampere University of Technology, Finland Jukka Huhtamäki, Tampere University of Technology, Finland Thomas Olsson, Tampere University of Technology, Finland Prof. Ravi Vatrapu, Copenhagen Business School, Denmark Prof. Hannu Kärkkäinen, Tampere University of Technology, Finland References Granovetter, Mark S. "The strength of weak ties." American journal of sociology 78.6 (1973): 1360-1380. Marsden, Peter V., and Karen E. Campbell. "Measuring tie strength." Soc. F. 63 (1984): 482. Gilbert, Eric, and Karrie Karahalios. "Predicting tie strength with social media." Proceedings of the SIGCHI conference on human factors in computing systems. ACM, 2009. Gupta, Jayesh Prakash, et al. "Identifying weak ties from publicly available social media data in an event." Proceedings of the 20th International Academic Mindtrek Conference. ACM, 2016. Availability of Big Social DATA Results • It is possible to identify potential weak ties from using the publically available social media data • Facebook data was not useful in identifying weak ties • Twitter was helpful in identifying the potential weak ties. • Temporal pattern of different social media channels use was different for Twitter and Facebook Limitations: • Insufficient number of survey respondents to draw any statistically significant result • Only some of the possible methods for weak tie identification were used Future Research • Combined other kind of data sources and methods for weak tie identification • Move towards utilizing Big Social Data • Utilize new analytics methods like Social Set Analysis • Validate the results using large number of survey respondents • The next steps: Please signup to be part of the future studies by providing your details on the following link or by scanning the QR code http://bit.ly/2jGOw64 Part of the COBWEB research project Funded by Academy of Finland (http://www.aka.fi). Presented Jan 23, 2017 at CMAD2017, University of Tampere, Tampere. Match Making in Professional Life Some Current Match Making Apps grip.events shapr.net Case CMAD 2016 Results Interaction networks Twitter Facebook Possible clusters based on Twitter Data Modularity class Cluster name 1 Personal branding 2 Employee advocacy 3 Drawings and infographics 5 **Broadly about cmadfi event 6 Community manager 7 Communications 10 *Reporting on CM ADFI event 15 Customer service 16 Project 17 *Outsider greetings 18 Tekes 20 Knowledge management 21 Jyväskylän energia Why is tie strength (weak and strong ties) important? Why social media / BSD analytics? Examples of matchmaking- related needs (which are rarely made use of e.g. in current matchmaking tools): Efficient spreading of information/research results/etc. to as large group of relevant people as possible (strong ties, authoritative/influencing persons) Identify new viewpoints and knowledge to problems or for new innovations / new interdisciplinary research (weak ties; both similar and new knowledge) Find relevant persons in professional events and conferences (find persons with novel / complementary expertise / persons with influence) Find interesting collaborators to a research project Different Kind of Ties • Strong ties: Strong ties are the ones that a person really trust. For eg. family • Weak ties: Refers to weak links which may be known.Eg. acquaintances, friend of friend • Weak ties are very useful in spreading novel information. New Possibilities with Big Social DATA Social Media Data , BSD and SNA (Social Network Analysis) in tie strength evaluation and matchmaking? Social Media data for example Facebook Friends, Twitter Follower/ Followee, have been used to identify the actual social network and tie strength of a person. Additionally, it is possible to use conversation data on social media to evaluate tie strength and help in identifying different kind of ties. SNA is one possible way to do this. SNA enables in easier identification of different influential people and the potential sub- communities in a network. BSD can e.g. enable using the historical conversation / other social data about conference/s to provide preliminary deductions about participants tie strength and the possible useful ties, their common or complimentary interests. For example co-occurrence of conference participants.