=Paper=
{{Paper
|id=Vol-1247/recsys14_poster9
|storemode=property
|title=Voting Operations for a Group Recommender System in a Distributed User Interface Environment
|pdfUrl=https://ceur-ws.org/Vol-1247/recsys14_poster9.pdf
|volume=Vol-1247
|dblpUrl=https://dblp.org/rec/conf/recsys/WorndlS14
}}
==Voting Operations for a Group Recommender System in a Distributed User Interface Environment==
https://ceur-ws.org/Vol-1247/recsys14_poster9.pdf
Voting Operations for a Group Recommender System in a
Distributed User Interface Environment
Wolfgang Wörndl Part Saelim
TU München TU München
Boltzmannstr. 3 Boltzmannstr. 3
85748 Garching 85748 Garching
Germany Germany
woerndl@in.tum.de part.saelim@tum.de
ABSTRACT allow users to rate items and remove recommended movies from
This work investigates distributed user interfaces for group the recommendation list, our solution proposes voting operations
recommender systems. In our scenario of a movie recommender, to help users agreeing on an item. We conducted a user study and
the user interface is distributed on two platforms: personal mobile present the results after explaining the main design features of our
devices and a public multi-touch tabletop. Our solution proposes application in this poster paper.
voting operations to better support the consensus building among
group members. We have implemented a prototype and conducted 2. SOLUTION DESIGN
a preliminary user study with interesting results. The main idea is to let users rate their movies on their personal
mobile device (1st subtask in the group recommender procedure)
Categories and Subject Descriptors and facilitate the result presentation and consensus building on a
H.5.3 [Information Interfaces and Presentation]: Group and multi-touch tabletop (3rd and 4th subtask). Users have the option
Organization Interfaces – Collaborative computing to browse movies individually and rate them on their mobile
device (Fig. 1), or bring existing ratings to the group session.
Users can move an item on the tabletop to a dedicated area of the
General Terms shared display, then the item details will appear on the personal
Design, Experimentation, Human Factors. mobile device.
Keywords
user interfaces, group recommender systems, multi-touch tabletop
1. MOTIVATION AND BACKGROUND
Recommender systems recommend items to an active user or user
group based on ratings of items or other information about users
and items. For group recommender systems, an additional
requirement is to facilitate the agreement process among group
member to select an item. For example, users may want to go to a
cinema together to watch a movie. They could rate some movies
on their personal mobile devices and the system could present
recommended items on a shared display such as a multi-touch
tabletop. The group can then use the tabletop to refine the query
and finally select a movie. A separation in the user interface is
also called "distributed user interface" (DUI). Thereby, the
components of the user interface are distributed across one or
more dimension (input, output, platform, space, and time) [1] with
multiple distribution strategies. Figure 1. User's view
Recommendation subtasks for group recommender consist of (1) On the tabletop, the system displays the results of the
acquiring preference, (2) generating recommendations, (3) recommendation process. The three best items are shown with a
presenting recommendations, and (4) helping group members to larger image and additional options are presented with smaller
arrive at an agreement [2]. Related work includes an early representation (Fig. 1). We refer to the differences in these
approach by McCarthy et al. [3]. The approach aims to support a representations as presentation modes. We use a simple
group of friends planning a skiing vacation. In our work, the goal aggregation algorithm based on maximizing average satisfaction
is to investigate distributed user interfaces and consensus building as recommendation algorithm. The tabletop provides an overview
for group recommender systems in more detail. In addition to of recommended items while more detailed information is
available on the mobile devices including explanations for
recommendations. In a group session, users can remove an item
from consideration for the current group by moving it to one of
the corners of the display. Users can also use a remove function
Copyright is held by the author/owner(s).
RecSys 2014 Poster Proceedings, October 6–10, 2014, Foster City,
on their mobile devices; in this case, an item is removed if more
Silicon Valley, USA. than half of group members submitted the removal.
In addition, we implemented several options to vote on the information might depend on the number people in a group
recommended items. The promote operation works like “thumbs- interacting with the system at the same time.
up” buttons commonly found in social networking sites. Similarly, remove requests submitted via the tabletop are more influential
our approach offers a demote option. Thus, group members can than via the mobile device, because only the former triggers the
express how they feel an item is suitable for the current group immediate removal of a movie. 10 out of the 21 participants
without having to alter their original personal rating for an item. agreed with this design choice. 29% of participants (6 of 21) felt
An important aspect is awareness: users should be able to see the that the system should treat the requests from both platforms the
opinions of other group members. This is implemented by smileys same way and 24% of participants (5 of 21) reported that the
(indicating actual or predicted ratings of other users) and thumbs- system should treat the requests from mobile devices as more
up/down (for demote/promote) icons above and below the movie influential than from tabletop. The other operations (promote,
images in our user interface (Fig. 1). The system also provides a demote or re-rate) were available on the personal mobile devices
re-rate operation on the mobile devices. re-rate can be used to only. So we asked the participants whether these operations
change the predicted or actual previous rating of an item, and should be provided on the tabletop as well. The results were
consequently affect the ranking of movies. mixed: 62% of participants felt that this alternative could be either
"very helpful" or "extremely helpful" for them. 33% of
3. USER STUDY participants felt that it could be either "not at all helpful" or
"slightly helpful" for them.
3.1 Setup
The participants were asked about the degree of usefulness of
We conducted a preliminary user study to evaluate our approach.
One of the objectives was to examine how recommendation each voting operation in terms of supporting them in arriving at a
final decision. For each operation, the response could be given on
results can be presented in the multi-device environment in order
a scale from 1 ("not at all useful") to 5 ("extremely useful"). The
to help a group of users to arrive at an agreement. We also wanted
re-rate operation received the highest mean score of 4.24, while
to investigate which functionalities the group recommender
system in this setup should offer on which device. promote and demote operations got mean scores of 3.67 and 3.62
respectively. The lowest mean score belongs to the remove
The first part of the experiment consisted of an explanation and operation of either platform, with a mean score of 3.43 for both
demonstration of all functionalities of the system to the options on the mobile device and the tabletop. But some
participants. Afterwards, the participants were given the task to participants noted that the promote and demote operations
determine what movie that want to watch together as a group by required lower effort to perform than to re-rate an item.
using the applications. All user interactions with the system were
Our findings indicate that the provided preference awareness
logged during the sessions. Finally, the participants were asked to
complete a survey about the user experience. plays an important role in helping the participants to arrive at an
agreement on a movie. Nearly half of the test users felt that the
To avoid cold start problems or a lengthy initiation procedure, we awareness was "remarkably useful" and none of them found it to
prepared five stereotype users with initials set of ratings. The be "useless". Some participants noted that they would even like to
users could choose one of the stereotypes and thus start the test receive notifications of other group members' activities on their
session with a reasonable user profile and some movie ratings respective mobile devices.
already present. Since evaluating recommendation accuracy was
The detailed information of movies is exclusively available on the
not one of the goals, this procedure allowed for a quicker setup.
mobile devices. However, to measure its usefulness, participants
The users were of course free to modify existing ratings or add
additional ones. were asked about how helpful the information was in terms of
facilitating the reaching of agreement on a movie. 14 participants
(67%) found the information to be either "very helpful” or
3.2 Results and Conclusion "extremely helpful". There were 2, 3 and 2 users who felt the
21 persons participated and were assigned to groups with at least information was "not at all helpful", "slightly helpful" and
3 users each. The sample consisted of 11 men and 10 women, "moderately helpful" respectively.
ranging between 20 and 30 years old. The participants were
mainly students and staff from our university. Interestingly, when the participants were asked to agree on a
movie at the end of the test session, all the groups decided on a
Overall, test users found our approach useful. 16 participants movie ranked as one of the top three movies in the group
(76%) found the prototype to be "very helpful" or "extremely recommendation list. Ongoing and future work focuses on the last
helpful" in supporting a user group to agree on a movie to watch. past of the introduced group recommender process: the support of
One more critical remark was that the number of recommended consensus building with refined alternatives for voting.
movies should be lower than our 15 proposed items in three
presentation modes. But the three different presentation modes
were evaluated as useful in principle.
4. REFERENCES
[1] Gallud, J.A., Tesoriero, R, and Penichet, V.M.E. 2011.
One question was whether the provision of information on items Distributed User Interfaces. Springer, Berlin/Heidelberg.
on both platforms (mobile/tabletop) is redundant. Most
[2] Jameson, A., and Smyth, B. 2007. Recommendation to
participants (71%) believe that it was not redundant and they
groups. The Adaptive Web, 596-627, Springer,
prefer to keep it available on both devices. 19% of participants (4
Berlin/Heidelberg.
of 21) would have preferred to access it on the mobile device only
and 10% of participants (2 of 21) preferred to access it on the [3] McCarthy, K. et al. 2006. CATS: A synchronous approach to
tabletop only. One participant remarked that the redundancy of collaborative group recommendation. Proc. FLAIRS
Conference, 86-91, AAAI Press