Local Friendship Programme – Privacy Statement

The both, the locals and the students use the same application form, which collects the following data:

• Names of the applicants
• Contact information
• Student's degree and the field of science
• Applicants' levels of fluency in the spoken language
• Hobbies and interests of the applicants
• Social profile, like a single, couple, family type, native Finnish, immigrant
• Gender identity (optional)
• Age in classification
• Possible pets and a wish to keep the meetings pet-free (allergies)
• Motivation to participate to the programme
• Things one appreciates in a friend
• Possible additional information for the LFP (optional)
• Critical criteria: two things that must be matched in profiles in order to accept the match results (optional)

Furthermore: Feedback received via the online feedback form

Data collection and data retention period

Personal data is collected directly from the data subject using an electronic form. Personal data is stored electronically in the university's secure storage facility, access to which is restricted to international services staff. The data is stored for one calendar year, after which it is destroyed for those individuals who do not continue in the program. Each participant has the right to request the deletion of their personal data at any time.

The results of the compatibility analysis are stored for as long as the pairing process lasts. After that, the data is deleted.

The description of a matching process

We comply with the University of Jyväskylä's artificial intelligence policy (15 April 2024) and artificial intelligence guidelines, particularly those relating to the use of MS Copilot AI.

First, we remove all direct identifiable information (names and contact details) from the raw data by recoding the data with identification codes such as SX and LX, where X is a number starting from 1. The identifying information is stored separately in secure disk space administered by the university, to which access is restricted. The identifying information is handled by designated personnel from the University of Jyväskylä's Student and Academic Services. In the analysis of pair options, we only use data from which all identifiable information has been removed.

We upload the data to the Microsoft Copilot artificial intelligence service via a VPN-protected and password-protected connection. The program helps us find the optimal set of SX-LX pairs. For this purpose, we use multi-objective optimization (MOO) tools in conjunction with qualitative text analysis.

The results consist of a summary; information about what the individuals have in common and what they do not have in common, without any details or identifiable information. Each pair suggestion is checked manually.

A summary of the suitable pair is sent without direct identifying information to the student and the local friend for their approval to the addresses provided in their application forms in blind copy format, where the recipients cannot see the email addresses of the recipients.

Once both parties have accepted the results, we will send the couple each other's contact details (names and email addresses only) by email. If both parties do not accept the proposal, we will cancel the pairing proposal and notify the parties, again anonymously. The participant's identification information will only be disclosed to the other person in connection with an accepted pairing proposal.