background image
Peer-to-peer based Recommendations for Mobile
Amund Tveit
Department of Computer and Information Science
Norwegian University of Science and Technology
N-7491 Trondheim, Norway
With the increasing number of mobile commerce facilities,
there are challenges in providing customers useful recom-
mendations about interesting products and services.
In this paper a Peer-to-Peer (P2P) based collaborative fil-
tering architecture for the support of product and service
recommendations for mobile customers is considered. Mo-
bile customers are represented by software assistant agents
that act like peers in the processing of recommendations.
Mobile commerce, or e-commerce in the wireless web, is pro-
viding commercial services that are accessible using mobile
devices, typically a mobile phone. The main advantage of
such services is their high availability, independent of phys-
ical location and time.
However, mobile commerce has its limitations, particularly
regarding communication and the resources of mobile de-
vices. Communication have less and more expensive band-
width, higher latency, lower connection stability, less pre-
dictability, and currently less standardized protocols [10].
Mobile devices in contrast to their desktop PC alternatives,
are severely constrained due to less processing and mem-
ory resources, smaller and less convenient user interfaces, in
addition to a limited energy supply.
An important question related to most types of commercial
activities is: Which products or services should be recom-
mended to a particular customer in order to make him/her
In e-commerce settings the automization of personalized rec-
ommendations has been sought solved using methods such
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise, to
republish, to post on servers or to redistribute to lists, requires prior specific
permission and/or a fee.
ACM Mobile Commerce Workshop 2001, July 21, 2001, Rome, Italy
Copyright 2001 ACM 0-12345-67-8/90/01 ...
as information filtering and collaborative filtering [4]. In-
formation filtering recommendations is based on the analy-
sis of a profile describing a (potential) customer's needs and
preferences. Collaborative filtering is based on using votes or
opinions about products and services from similar customers
in order to give recommendations. Collaborative filtering
has been shown as the best approach of these two methods,
but it is highly dependent on a large number of customers
to give good recommendations and it doesn't scale well in
terms of processing.
In this paper an approach for making a scalable recommen-
dation system for mobile commerce using a Peer-to-Peer
(P2P) is considered. Peers are represented as software as-
sistant agents interfacing a mobile customer. This approach
adds the opportunity of product and service recommenda-
tions to the mobile commerce agent architecture presented
in [6].
Customer 1
Agent 1
Agent n
Customer n
Agent 1
Agent k
Figure 1: Mobile Commerce Agents
The rest of the paper is organized as follows. Section 2 de-
scribes collaborative filtering theory. Section 3 describes the
proposed approach for P2P-based recommendations. Sec-
tion 4 describes related work, and finally the conclusion with
future work.
P2P-based Recommendations for M-Commerce

<< - < - > - >>