Number of Visitors: 1513
 

:

Master Thesis Defense Session

Master Thesis Defense Session by Mahshid Keivandarian, Personalization of Explanation Style in Persuasive Recommender Systems Using Psychological Information of Users Extracted from Social Network

News Code: 15

Publishing Date: 11 Sep 2022 16:4



In The Name of God

Master Thesis Defense Session

Computer Engineering, IT Engineering, Electronic Commerce​

 

Supervisor:

Dr. Fakhrodin Noorbehbahani

Dr. Azadeh Mohammadi

Internal Reviewer:

Dr. Marjan Kaedi

External Reviewer:

Dr. Afsaneh Fatemi

Researcher:

Mahshid Keivandarian

Date: 12 September 2022

Time: 9:30 AM

Location:

Ansari building, Third floor, Dr. Braani Hall

Online link : lms.ui.ac.ir

 

Guest Account:

Username: computer

Password: computer1305

 

Topic:

Personalization of Explanation Style in Persuasive Recommender Systems Using Psychological Information of Users Extracted from Social Network

Most IoT devices operating in mobile computing environments have networks with variable error rates andPersuasive recommender systems emerged as a new generation of recommender systems in response to the challenge of users' lack of motivation to purchase products or services suggested by conventional recommender systems. In addition to the components of a normal recommender system, these systems also need persuasive elements to persuade users. Studying persuasion elements is very important in terms of performance improvement in these systems in both academic and business respects. Due to the different psychological nature of users, persuasive explanations should also be personalized to be more effective, and this issue is quite challenging due to the high diversity and complexity of human nature. The proposed method of this research is to personalize the style of presenting explanations in persuasive recommender systems using psychological information obtained from users' social networks. According to the results, effective persuasion strategies on users can be obtained by using psychological and demographic information collected from them. In addition, in order to prevent any sort of bias of users leading to resistance to persuasion, their psychological information was obtained indirectly. This thesis has both academic and business aspects since it can be implemented in any type of business using recommender systems and as a base reference in further studies.