Abstract
dc:description.abstractPeople’s online content choices should be driven by their intentions, but can be greatly affected by online recommendation systems. Existing recommendation systems mostly focus on promoting popular items and users’ historical records, thus may bias users against their original intentions. To study the problem, we launched a field study to compare the effects of intention-informed recommendations with classical intention-agnostic systems in the context of YouTube videos. We built a website to collect participants' intended video categories, recall video categories, and their opinions and usage of YouTube. We aim at conducting an intervention study with different recommendation on control and test group. The intervention is conducted with automatic emails containing videos of different categories, either intended categories suggested by the participants or the categories recalled by them. By examining users' responses under different interventions, we could compare the effects of intention-aware and unaware recommendations. As a reward for our participants, we also build an interactive visualization platform for people to know better about their own YouTube online data.
Degree
thesis:*- Name thesis:degree_name
- M.S., Information Science
- Level thesis:degree_level
- Master of Science
- Discipline thesis:degree_discipline
- Information Science
- Grantor
- Cornell University
- Year dc:date.issued
- 2020
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Wang, Yixue
- Yao, Siyu
- Chair dc:contributor.chair
-
- Azenkot, Shiri
- Committee member dc:contributor.committeemember
-
- Estrin, Deborah
Subjects
dc:subject × 2Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Bibid
- 13254479
- Dc Identifier Other
-
ProQuest Submission ID: 10853
ProQuest Publication ID: 27957758
ProQuest Submission ID: 10812
ProQuest Publication ID: 27834177