=Paper= {{Paper |id=Vol-1664/w11 |storemode=property |title=A Group Catalog Mechanism for Promoting Knowledge Sharing in Open Virtual Communities |pdfUrl=https://ceur-ws.org/Vol-1664/w11.pdf |volume=Vol-1664 |authors=Fabrizio Messina,Francesco Alessandro Sarné |dblpUrl=https://dblp.org/rec/conf/woa/MessinaS16 }} ==A Group Catalog Mechanism for Promoting Knowledge Sharing in Open Virtual Communities== https://ceur-ws.org/Vol-1664/w11.pdf

A Group Catalog Mechanism to Promote Knowledge
Sharing in Open Virtual Communities
Fabrizio Messina and Francesco A. Sarné

Abstract—In open virtual communities, thematic groups pro- associated agents) will result improved. Therefore, it is conve-
mote mutual cooperation among their members in order to reach nient to support users’ interactions within a virtual community
specific targets. To this purpose, users share portions of their (i.e. a thematic group) by means of some mechanisms capable
knowledge in a reciprocal understandable manner. For this aim, to provide a suitable representation of personal knowledges in
personal software agents are able to assist users by encoding a mutually understandable manner.
personal information about preferences and goals into suitable
Given the premises above, we propose to adopt a spe-
profiles. In this work we present a multi-agent solution to manage
knowledge shared by users across a number of common thematic cialized thematic catalogue storing topics (i.e. names, things,
groups. A common catalog is created for each thematic group of concepts and so on) of interest in thematic groups in order
interest that, in turn, is associated with a group agent. The group to provide potentially heterogeneous agents with a mutually
agent is devoted to support the group by interacting with personal understanding common knowledge. The catalog is publicly
agents in order to manage the group affiliation process and enrich available by all the agents affiliated with that thematic group
the common thematic catalog of its own group. In presence of and represents the common knowledge with respect to all
heterogeneous agents, such a common group catalog is a key the topics dealt within that group. At the same time, users’
element to provide knowledge sharing and agent interoperability agents are provided with individual knowledge deriving by
with both other personal and the group agents. In the proposed the analysis of both the past and the current behaviors of their
approach each user agent is able to personalize its own catalog
users. In order to include individual knowledge, the common
and contribute to enrich that of its own group by collaborating
with its group agent. shared catalog of a thematic group can be enriched by means
of the mutual cooperation between the users’ agents affiliated
Keywords—Open Virtual Communities, Knowledge Sharing, with that community and the group agent managing it which,
Common Thematic Catalog, Intelligent Agents, Thematic Groups. periodically, provides to update such a common catalog.
The remainder of the paper is as follows. Section II contains
the reference scenario, while in Section III we discuss the
I. I NTRODUCTION
structure of the designed catalogue. In Sections IV and V the
In open virtual communities [1]–[5] users having an interest profiles of the personal and the platform agents are described.
for a common topic (e.g. sports, food) look for a profitable Section VI presents some related literature and the novelties
opportunity to collaborate in order to satisfy their needs. In provided by this work. Finally, in Section VII we draw our
such environments a common way to promote these activities conclusions and introduce our future works.
consists of creating thematic groups formed by users sharing
common interests. II. T HE O PEN M ULTI -AGENT A RCHITECTURE
To maximize the quality of interactions within each thematic The proposed model considers a number of thematic groups
group, software agents may be employed to assist users [6]– within several open Virtual Communities (V ), each one spe-
[9], in order to carry out important tasks related to knowledge cialized on a specific topic or set of topics (hereafter only
sharing that may result heavy and boring [10]. Each software topic). Furthermore, each thematic group can affiliate users
agent is able to build a personal profile for its own user belonging to different open virtual communities and each user,
by monitoring the user’s activities carried out within the in turn, can be member of different thematic groups, each one
community. Therefore, every thematic group will correspond to potentially belonging to a different open virtual community.
a group of software agents on which a group agent will manage Each user u is supported by a software agent a, called
the group itself. In this context, software agents shall adopt Personal Agent, which is specialized on the topic t charac-
different descriptions to describe the same topic of interest. terizing a specific group G. Therefore, when a user is joined
When a common representation of the users’ knowledge is with more thematic groups, he/she will be supported by a set
not available, then such interactions among users (i.e. agents) of Personal Agents, one for each group (i.e. topic). In order
could be not possible. Conversely, when a representation to support its owner in performing his/her activities within a
of knowledge which results mutually understandable, quality group, his/her Personal Agent suitably encodes in its profile
relationships and cooperations among users (i.e. among the all the information necessary to manage the user’s interests
for the specific topic of that group. Similarly, each group is
Fabrizio Messina is with the Dept. DMI, University of Catania, Viale Andrea managed by the Group Agent A devoted to provide some
Doria, 6 - 01010 Catania, Italy, e-mail: messina@dmi.unict.it
Francesco A. Sarné is with the Politecnico di Milano, Piazza basic services to its affiliated users by cooperating with their
Leonardo da Vinci, 32 - 20133 Milano, Italy, e-mail: francescoalessan- associated Personal Agents. The proposed model architecture
dro.sarne@mail.polimi.it is graphically depicted in Figure 1.

62
2

Open Virtual Community V 1
Group Agents
A1 A n-1 An
GROUP 1 GROUP N-1 GROUP N

Topic 1 Topic N-1 Topic N

Personal Agents

Users

Open Virtual Community V 2 Open Virtual Community V 3
Fig. 1. The proposed Open Virtual Community architecture.

Given the open nature of the proposed architecture, Personal thematic group. As described above, each of these catalogs
Agents coming from different virtual communities could en- is associated with a thematic group and shared among all
code their knowledge with different modalities. This implies its members and stores all those topics (i.e. names, things,
that different agents may represent the same topic by using concepts and so on) and their mutual relationships resulting of
different terms as well as the relationships linking it to other interest for the group members. We assume that the catalog
topics may be different between two agents. Consequently, C contains the common knowledge of a group. It is publicly
in order to promote a better reciprocal understanding it is available to all the members of this group and each Personal
necessary to provide each thematic group with some suitable Agent can enrich it with further knowledge, also including any
mechanism in order to give a common knowledge, specialized new relationships among the new entries and the past common
for that group, to the agents. knowledge.
To this purpose, in this paper for each thematic group it is
proposed the adoption of a Thematic Catalog (C) storing all the A. Components of C
topics and their mutual relationships which form the common
knowledge for all the agents affiliated with that thematic A Thematic Catalog C consists of a set of topics i) Ctopic ii)
group. Such a Thematic Catalog is publicly available at all a set Clink of links which represents the relationships existing
the Personal Agents of a thematic group and it is periodically among the topics belonging to Ctopic . A link between the
updated by the associated Group Agent. Therefore, Personal two topics ti , tj ∈ C is described by a tuple in the form of
Agents in managing their users profiles, in order to mutually hti , tj , Li,j , pi,j i where:
cooperate with the other members of their group, can directly • ti and tj are the topics which are identified by two
represent the topics of interest, already present in C, by using lexical terms that are linked in the thematic catalog C;
the corresponding terms associated in C. Moreover, Personal • Li,j identifies the type of the link involved in the
Agents, by monitoring their users, can acquire new knowledge relationship occurring between ti and tj ;
referred to a thematic group (i.e., those topics currently not • pi,j is a parameter giving information on some char-
belonging to C) and can use C also to represent it into acteristics of the link, by means of a numerical value
their personal knowledge by means of a general relationship ranging in [0, 1] ∈ R.
between each “personal” topic with at least another topic More in detail, we introduce two types of category links,
already present in C to allow the interoperability among all namely:
the agents of the same group also on such personal topics. • I : this type of link connects two topics ti and tj iff
the terms belonging to ti have a different meaning of
III. T HE T HEMATIC C ATALOG those belonging to tj (for instance, the terms painting
and sculpture). In this case, the value of pi,j represents
This section provides a formal description of the Thematic the degree of interest that a user interested in ti has
Catalog (C) which permits to the Personal Agents of inter- about tj which can vary from null (i.e, pi,j = 0.0) to
acting with the other Personal Agents affiliated to the same maximum (i.e., pi,j = 1.0).

63
3

• II : this type of link connects two topics ti and tj Moreover, each topic t ∈ C of interest for uk is associated
which can belong to three categories based on the PC
in Tcatk with a tuple Γk = hik , vk i, where i ∈ [0, 1] ⊂ R
value assumed by the parameter p, namely (i) isa, (ii) represents the interest of uk for t (respectively, 0/1 denotes
overlapped or (iii) synonymous. In particular, for the the minimum/maximum interest for t), while v is a flag which
three considered categories we have that: specifies the type of visibility that uk desires to give to his/her
◦ isa, iff the terms belonging to ti also belong to own interest for t (respectively the value 0/1 corresponds to a
tj and in this case pi,j = 0.0. For instance, with public or private visibility). Figure 3 reports an example of
respect to the terms bust (ti ) and sculpture (tj ) it agent profile derived from the example proposed in Figure 2.
means that each bust is also a sculpture. Note the categories (in bold) are not present into figure 2.
◦ overlapped, iff some terms of t1 also belong to Determining topics of interest. The Personal Agent ak,i
t2 and vice versa, in this case pi,j ranges in the monitors the activities of its user uk within the thematic group
domain ]0, 1[. For instance, the two terms cup and Gi and periodically provides to evaluate the interest of uk in
goblet are partially synonymous because a cup the topics belonging to its profile PkCi . To this purpose, for
could not be exactly a goblet and vice versa. each topic in the profile PkCi the agent ak , by collaborating
◦ synonymous, iff the terms belonging to ti have with the other agents which uk belongs to and from which it
the same meaning of those belonging to tj , in this collects all their catalogs, computes the index Ik,t as:
case pi,j = 1.0. For instance, the terms statue and
sculpture identify the same type of artistic artifact.
1
Ik,t = P (1)
B. Representation of the Thematic Catalog C ∀t∈{Sk ∩Cuk } it
−
A Thematic Catolog C is representable by using a direct 1+e kSk ∩ Cuk k
graph T C = hTtopic C C
, Tlink C
i, where Ttopic is the set of nodes,
C
each one associated with a different topic. Similarly, Tlink
represents the set of arcs of the graph, where each link in where Sk = {t1 , t2 , . . . , tn } is a set of topics of interest for
C
Tlink is associated with a relationship hti , tj , Li,j , pi,j i ∈ C, uk that belong to PkC and Cuk is the set of catalogs of all the
with L ∈ {I, II} and pi,j ∈ [0, 1] ∈ R, as explained in groups where uk is member. For each topic t belonging to Sk
Section III-A. In the following of this paper, we will refer is determined the average interest shown by uk (i.e., Ik,t ) with
to the Thematic Catalog as T C or C in an interchangeable respect to the domain [0, 1]. More in detail, when the interest
manner. Figure 2 shows an example of a Thematic Catalog C of the user uk in a topic t decreases then, consequently, the
concerning Art. In particular, links that belong to more than value of the associated interest it for that topic decreases and,
one category have multiple labels. Note that in Figure 2 all the as a result, also the value of Ik,t will decrease. Conversely, for
arcs are depicted without orientation but, in order to take into any group concerning the same topic managed by the agent
account the different possible characteristics of the links, for ak , their computed values of Ik,t will increase.
convenience both the links of type I when pi,j = 1.0 (i.e., the Group affiliation. Based on a threshold ψ ∈ [0, 1] ∈ R
two topics are not disjointed) and of type II when pi,j = 0.0 fixed by the user, each Personal Agent provides to identify
(i.e., the topics are isa) are depicted as oriented. those groups potentially of interest for its own user as well
Moreover, we define that two categories ti and tj as to require the affiliation to their respective Group Agents.
are in a t-relationship when in C there exists a path Similarly, the Personal Agent also suggests to leave a group as
hti , tk , Li,k , pi,k i . . . htm , tj , Lm,j , pm,j i. Differently, if the well as to send a leave message to the Group Agents managing
links of this path joining the nodes ti and tj belong to different those groups for which the interest of its user is low.
topics links, we say that they are generally related. More in detail, with respect to the Personal Agent ak,i after
that the index Ik,t has been computed, when Ik,t > ψ for a
IV. T HE P ERSONAL AGENT topic if interest for uk , if there is a group focused on that topic
then it becomes a candidate for the user to join with (and a
Each user uk is assisted by several Personal Agents new Personal Agent of uk could be activated to deal with all
(ak,1 , ak,2 , . . . , ak,n ), where n is the number of groups to the activities of uk within this group). At the same way, when
which the user uk is affiliated (i.e. topics of interest for uk ). for a group for a topic of interest it results that Ik,t < ψ, if uk
More in detail, for each group where a user is affiliated his/her is affiliated with a group focused on that topic then its Personal
associated Personal Agent will manage his/her affiliation with Agent recommends to the user of leaving that group and, if
that group (and with other groups focused on the same topic). required by its user, it provides to send a leave request to the
User Profile. The Personal Agent ak monitors all the Group Agent associated with it and the associated Personal
activities of its own user uk referred to a specific group in Agent will be stopped.
order to maintain the user’s profile PkC . To represent the profile Matching category links. Another activity is executed by
PkC for the user uk , the same notation of C is adopted, i.e. a the Personal Agent on existing connections between topics
C PkC PkC PkC t1 ∈ {Sk ∩Cuk } and t2 ∈ {Sk −Cuk }, with respect to the links
graph T Pk = hTtopic , Tlink i where Ttopic is the set of topics of type II — i.e., synonymy (s) and overlap (o). In particular,
PC
and Tlink
k
is the set of links. the parameter Hk is computed as:

64
4

II(o)
Affresco Bronze Manzù
Oil II (s,o) II (i)
II (s,o)

II (i)
Watercolor Paint Art Sculpture II(o) II(o) I

Crayon II (i) II (i) II(s,o)

Bust Milan

Fig. 2. A part of a CTD about “Art”. Note that for the type II, the links isa, synonymous, overlap have been respectively identified as i, s and o.

X of all the group members. A third data structure, named Yellow
1 Pages, is devoted to store all the public interests of the group
Hk (t1 , t2 ) = · vtm1 ,t2 · N m (t1 , t2 )
Nt2 members in order to allow each agent to find in the group
m∈{s,o}
other agents (i.e. users) sharing similar interests in the same
where t1 ∈ {Sk ∩ Cuk }, t2 ∈ {Sk − Cuk }, vtm1 ,t2 represent topics. The Yellow Pages data structure is formed by a set of
the parameters vts1 ,t2 and vto1 ,t2 ∈ [0, 1] ⊂ R as well as Ntm 1 ,t2
lists, each one referred to a single agent (i.e. user) resulting
represents N s (t1 , t2 ) and N o (t1 , t2 ), which are the number affiliated with the group.
of synonymy and overlap connections between t1 and t2 ,
respectively. Moreover, Nc2 is the total number of links of the A. Group Agent behavior
topic t2 ∈ {Sk −Cuk } and since N s (t1 , t2 )+N o (t1 , t2 ) ≤ Nt2 ,
it will be Hk < 1. Group affiliation. When a user joins with the group assisted
The purpose of the computation of Hk,Cj is to select those by the Group Agent, he/she receives an identifier for that group
topics belonging to users catalogs, in order to enrich the and, consequently, the White pages of the group are updated.
catalog of the group with further users categories, as explained Similarly, when a user leaves the group then its associated
in the following. Group Agent will prune all the information referred to that
Catalog enrichment. Let be j ∗ a group to which user uk user from his/her data structure.
is affiliated. Firstly, the uk Personal Agent ak,j ∗ provides to Dictionary enrichment. The Thematic Catalog of a group
calculates the value Hk for all t1 ∈ {Sk ∩ Cuk } and t2 ∈ is periodically updated by the associated Group Agent basing
{Sk − Cuk }. After this, the agent ak,j ∗ calculates the index on the knowledge of the affiliated Personal Agents (see Sec-
Hb k (t2 ) ∀t2 ∈ {Sk − Cu } as: tion IV). Indeed, the Group Agent will collect all the topics
k
t 6∈ Cuk sent to it by Personal Agents affiliated with its group
1 X because H b k is greater then the φ parameters set by their owners
b k (t2 ) =
H Hk (t, t2)
||Sk ∩ Cuk || (see Section IV). Note that when an agent is interested to
t∈Sk ∩Cuk enrich the catalog of its group with a new topic t it is “quite
Moreover, the Personal Agent ak,j ∗ sets the threshold φ ∈ connected” with other categories belonging to the set of topics
b k (t2 ) ≥ φ, then the topic t2 ∈ {Sk − Suk . The information Hk is for the Group Agent a first set of
[0, 1] ⊂ R such that it is H
candidate topics from which it will extract those topics having
Cuk } is sent to the group agent in order to enrich the catalog
the highest frequency f 1 or, in other words, those sent by
of the group. When the Group Agent receives the topic t2 then
the higher number of Personal Agents in order to use that
it will be a potential candidate to be added to the catalog of
knowledge resulting really shared among the groups members.
own group by means of the selection of other parameters, as
the frequency of the involved terms (see Section V).
VI. R ELATED W ORK
V. T HE G ROUP AGENT A wide body of studies investigated on the various modali-
ties to promote interactions and mutual cooperation in hetero-
geneous environments [12]–[14]. Consequently, in this section
Group Agents are the counterparts of the Personal Agents only those approaches which comes closed to the arguments
(i.e. users) with respect to the activities of group affiliation proposed in this paper will be cited. The interested reader can
and enrichment of the Thematic Catalog C management. To refer to [15]–[18] for a more complete overview on the matter.
support such activities the Group Agent adopts specific data The capability of a mutual collaboration among agents
structures able to encode the group profile. [11] usually implies a mutual understanding ability, and in this
More in detail, (see Section III), the Thematic Catalog C context a common way is that of providing agents with some
of the group stores all those topics of interest for the group form of knowledge shared by all the cooperating agents [19].
members, as well as all the relationships taking place among
them (see Section III). A White Pages service is provided to the 1 The frequency is computed among all the used terms which fall into these
agent affiliated with the group in order to provide the identifiers topics

65
5

I
I Surace Affresco Bronze Manzù
Oil II (s,o) II (i)
II (s,o)
I
II (i)
Watercolor Paint Art Sculpture II(o) II(o) I

II (i)
Crayon II (i) II (i) II(s,o)

Photography Bust Milan
I

I
II(o) II (i)
Lens Nikon Nikon D800

Fig. 3. An example of Personal Agent profile based on the Catalog of Figure 2 (new topics are represented in bold and new links are represented with dashed
lines).

A similar approach has been adopted in [1] where agents mantic communications. Another technique is presented in [28]
share a common hierarchical ontology representing a close and consists of using shared keys, which are semantically
domain of interest. In fact, in this proposal the agents have not negotiated by agents, to solve the problem deriving by the
the opportunity to represent their individual knowledge and, presence of synonymies in order to avoid the adoption of
therefore, in a more general context it results highly limited different terms for the same objects and, in this way, permitting
to support emerging user’s needs. the mutual agent understanding.
Other approaches where a common and shared ontology Finally, in [29] and [30] users are supported by a set of
is unnecessary are proposed in [20]–[23]. More specifically, personal agents. More in detail, in a benevolent environment,
the authors of [20] in a message-based mechanism propose a they adopt an approach inspired to the biologic evolution where
meta-ontology for translating the presuppositions extracted by the best performing agents can be cloned and the worst agent
a message in order to make understandable its meaning to the can be deleted. Similarly to the proposal presented here, each
receiver agent. This target is obtained by means of a common user is supported by more agents, potentially heterogeneous for
vocabulary shared between the sender and the receiver agents. knowledge representation modalities, that, differently from this
Moreover, in presence of conflicts, inconsistencies or onto- proposal, act autonomously and therefore they do not need to
logical gaps in the incoming message then the receiver agent communicate. However, a similar approach could receive great
has the possibility to change its personal ontology in order to benefits from the adoption of a common catalog which can be
overcome such problems, while other systems have chosen to enriched by the individual and potentially heterogeneous agent
adopt semantic negotiation approaches as in [14], [24] knowledges, although these proposals implement a not explicit
In a similar way, the paper [22] presents a domain-specific knowledge sharing on the basis of the cloning process. In this
ontology, called global ontology, which allows a matchmaking overview we presented some approaches implementing mutual
system to be used by agents. This approach gives the advantage collaboration among software agents on the basis of their
of not adopting shared ontologies. More precisely, each agent mutual understanding and other implementing a multiple agent
provides to its platform the map of its ontology which is support for each user. The most part of them implement some
integrated in that of the platform. This task is executed by form of shared knowledge as dictionaries, common/global
using a suitable extraction engine which provides to identify ontologies more or less versatile.
relevant information present into the personal agent ontologies. In particular, our proposal is based on a dictionary approach
Therefore, the aim of the common ontology shared on the to promote agent cooperation in a simple and versatile way.
platform is only that of a dictionary for translation and Indeed, the proposed catalog natively permits to the agent of
each agent can use its personal conventions. Other solutions enriching it in order to include both the common and the
adoptable when it is needed to represent wide and specialized personal knowledge which give to the groups the dynamic
knowledge contexts, as in the e-Commerce, are those adopted capability of easily evolving. Currently, a prototype of a
in [25] and [26] where the interacting agents have to perform framework based on the proposal presented in this paper is in
the task of realizing their mutual understanding by allowing an implementation phase in order to test its real performance.
them the capability to build rich and detailed XML users’
profiles.
The authors of [27] studied the problem of the potential VII. C ONCLUSIONS AND FUTURE WORK
heterogeneities existing in digital libraries. To this purpose This paper discussed the problem to promote mutual users
they designed a P2P agent framework by associating each interactions and cooperation within thematic groups in open
library with a software agent aimed to realize a common virtual (agent) communities in presence of heterogeneous
dictionary (i.e. ontology) capable to support the agents in se- knowledges among the affiliated users (i.e. the associated

66
6

agents). To this aim, a framework is proposed such that each
[13] P. De Meo, F. Messina, D. Rosaci, and G. M. Sarné, “Recommending
thematic group is assisted by a Group Agent and, in turn, users in social networks by integrating local and global reputation,”
each user is assisted by a Personal Agent. More in detail, each in International Conference on Internet and Distributed Computing
Personal Agent is specialized only on a specific theme (i.e. Systems. Springer International Publishing, 2014, pp. 437–446.
topic) and manages a personal profile (resp. catalog) of its [14] A. Comi, L. Fotia, F. Messina, G. Pappalardo, D. Rosaci, and G. M. L.
owner’s knowledge and interests, such that users are supported Sarné, “Using semantic negotiation for ontology enrichment in e-
by one or more Personal Agents. In such a context, Group learning multi-agent systems,” in Complex, Intelligent, and Software
Agents provide to their affiliated Personal Agents some basic Intensive Systems (CISIS), 2015 Ninth International Conference on.
IEEE, 2015, pp. 474–479.
services. Each group catalog is extensible by the delegated
Personal Agent in order to take into account other topics [15] O. Hiroyuki et al., “A unified view of heterogeneous agents’ interac-
tion,” IEICE TRANSACTIONS on Information and Systems, vol. 84,
of interest for its user. Then such further knowledge can no. 8, pp. 945–956, 2001.
be exploited by the Group Agents to enrich their respective
[16] B. Chaib and F. Dignum, “Trends in agent communication language,”
common Thematic Catalogs of their groups. As future work, Computational intelligence, vol. 18, no. 2, 2002.
we will perform a number of simulations in order to verify the
effectiveness of this proposal. [17] C. Antonelli, “Models of knowledge and systems of governance,”
Journal of Institutional Economics, vol. 1, no. 01, pp. 51–73, 2005.
[18] S. Costantini and G. Gasperis, “Exchanging data and ontological
R EFERENCES definitions in multi-agent-contexts systems,” in RuleMLChallenge track,
[1] F. Buccafurri, D. Rosaci, G. M. L. Sarné, and L. Palopoli, “Modeling Proceedings. CEUR Workshop Proceedings, CEUR-WS. org, 2015.
cooperation in multi-agent communities,” Cognitive Systems Research, [19] J. Waters, B. J. Powers, and M. G. Ceruti, “Global interoperability
vol. 5, no. 3, pp. 171–190, 2004. using semantics, standards, science and technology (gis 3 t),” Computer
[2] S. Gauch, M. Speretta, A. Chandramouli, and A. Micarelli, “User Standards & Interfaces, vol. 31, no. 6, pp. 1158–1166, 2009.
profiles for personalized information access,” in The Adaptive Web, ser.
[20] R. Beun, R. van Eijk, and H. Prust, “Ontological Feedback in Multiagent
LNCS, vol. 4321. Springer, 2007, pp. 54–89.
Systems,” in AAMAS ’04: Proceedings of the 3rd International Joint
[3] P. De Meo, F. Messina, D. Rosaci, and G. M. Sarné, “Improving Conference on Autonomous Agents and Multiagent Systems. Washing-
the compactness in social network thematic groups by exploiting a ton, DC, U: IEEE Computer Society, 2004, pp. 110–117.
multi-dimensional user-to-group matching algorithm,” in Intelligent
Networking and Collaborative Systems (INCoS), 2014 International [21] L. Palopoli, D. Rosaci, and G. M. L. Sarnè, “Introducing specialization
Conference on. IEEE, 2014, pp. 57–64. in e-commerce recommender systems,” Concurrent Engineering, p.
1063293X13493915, 2013.
[4] P. De Meo, F. Messina, G. Pappalardo, D. Rosaci, and G. M. Sarnè,
“Similarity and trust to form groups in online social networks,” in OTM [22] D. Embley, “Toward Semantic Understanding: An Approach Based
Confederated International Conferences” On the Move to Meaningful on Information Extraction Ontologies,” in CRPIT 04: Proceedings of
Internet Systems”. Springer International Publishing, 2015, pp. 57–75. the 15th Australasian Database Conference, Volume 27. Australian
[5] A. Comi, L. Fotia, F. Messina, G. Pappalardo, D. Rosaci, and G. M. Computer Society, 2004, pp. 3–12.
Sarné, “Forming homogeneous classes for e-learning in a social net- [23] D. Rosaci and G. M. L. Sarnè, “Multi-agent technology and ontologies
work scenario,” in Intelligent Distributed Computing IX. Springer to support personalization in b2c e-commerce,” Electronic Commerce
International Publishing, 2016, pp. 131–141. Research and Applications, vol. 13, no. 1, pp. 13–23, 2014.
[6] M. Wooldridge and N. R. Jennings, “Intelligent agents: Theory and
[24] F. Messina, G. Pappalardo, D. Rosaci, and G. M. L. Sarnè, “An
practice,” The knowledge engineering review, vol. 10, no. 02, pp. 115–
agent based negotiation protocol for cloud service level agreements,”
152, 1995.
in Enabling Technologies: Infrastructure for Collaborative Enterprise,
[7] F. Messina, G. Pappalardo, D. Rosaci, C. Santoro, and G. M. L. Sarné, 2014. 23th IEEE International Workshops on. IEEE, 2014, pp. 161–
“A trust model for competitive cloud federations,” Complex, Intelligent, 166.
and Software Intensive Systems (CISIS), pp. 469–474, 2014.
[25] P. De Meo, D. Rosaci, G. Sarnè, G. Terracina, and D. Ursino, “EC-
[8] F. Messina, G. Pappalardo, D. Rosaci, and G. M. Sarné, “An agent
XAMAS: Supporting E-Commerce Activities by an XML-based Adap-
based architecture for vm software tracking in cloud federations,” in
tive Multi-Agent System,” Applied Artifificial Intelligence, vol. 21,
Complex, Intelligent and Software Intensive Systems (CISIS), 2014
no. 6, pp. 529–562, 2007.
Eighth International Conference on. IEEE, 2014, pp. 463–468.
[9] P. De Meo, F. Messina, D. Rosaci, and G. M. Sarné, “An agent- [26] D. Rosaci, G. M. L. Sarnè, and D. Ursino, “A multi-agent model
oriented, trust-aware approach to improve the qos in dynamic grid for handling e-commerce activities,” in Database Engineering and
federations,” Concurrency and Computation: Practice and Experience, Applications Symposium, 2002. Proceedings. International. IEEE,
vol. 27, no. 17, pp. 5411–5435, 2015. 2002, pp. 202–211.
[10] A. Comi, L. Fotia, F. Messina, G. Pappalardo, D. Rosaci, and G. M. [27] H. Ding and I. Sølvberg, “Towards the Schema Heterogeneity in
Sarné, “Supporting knowledge sharing in heterogeneous social network Distributed Digital Libraries,” in ICEIS (5), 2004, pp. 307–312.
thematic groups,” in Complex, Intelligent, and Software Intensive Sys-
[28] R. Guha, “Semantic Negotiation: Co-identifying Objects Across Data
tems (CISIS), 2015 Ninth International Conference on. IEEE, 2015,
Sources,” in AAAI ’04 Spring Symposium Series: Proceedings of the
pp. 480–485.
Semantic Web Services, March 2004.
[11] P. De Meo, E. Ferrara, D. Rosaci, and G. M. L. Sarné, “Trust and
compactness in social network groups,” Cybernetics, IEEE Transactions [29] D. Rosaci and G. M. L. Sarnè, “EVA: an evolutionary approach to
on, vol. 45, no. 2, pp. 205–216, Feb 2015. mutual monitoring of learning information agents,” Applied Artificial
Intelligence, vol. 25, no. 5, pp. 341–361, 2011.
[12] P. De Meo, F. Messina, D. Rosaci, and G. M. L. Sarné, “2d-
socialnetworks: Away to virally distribute popular information avoiding [30] ——, “Cloning mechanisms to improve agent performances,” Journal
spam,” in Intelligent Distributed Computing VIII. Springer Interna- of Network and Computer Applications, vol. 36, no. 1, pp. 402–408,
tional Publishing, 2015, pp. 369–375. 2012.