A Peer-to-Peer Architecture for Distributed Knowledge Management.

UNIVERSITY
OF TRENTO
DEPARTMENT OF INFORMATION AND COMMUNICATION TECHNOLOGY
38050 Povo – Trento (Italy), Via Sommarive 14
http://www.dit.unitn.it
A PEER-TO-PEER ARCHITECTURE
FOR DISTRIBUTED KNOWLEDGE MANAGEMENT
Matteo Bonifacio, Roberta Cuel,
Gianluca Mameli and Michele Nori
2002
Technical Report # DIT-02-0065
.
A Peer-to -Peer Architecture
for Distributed Knowledge Management
M a t t e o B o n i f a c i o 1 , R o b e r t a C u e l 2 , G i a n l u c a M a m e l i 3 , Michele Nori 3
1
University of Trento, Department of Information and Communication Technology
Via Sommarive, 18 -- 38050 Trento, Italy
[email protected]
2
University of Trento, Department of Computer and Management Science,
Via Inama, 5 -- 38100 Trento, Italy
[email protected]
3
ITC-IRST, Automated Reasoning Systems Division,
via Sommarive, 18 -- 38050 Trento, Italy
{mameli, nori}@irst.itc.it
Abstract. Most of the knowledge management systems of complex organizations are based on technological architectures that are in contradiction with the social processes of knowledge creation. In particular, centralized architectures are
adopted to manage a process that is intrinsically distributed. In this paper, assuming a Distributed approach to Knowledge Management (DKM), is proposed that
technological and social architectures must be reciprocally consistent. Moreover,
in the domain of Knowledge Management, technological architectures should be
designed in order to support the interplay between two qualitatively different
processes: the autonomous management of knowledge of individuals and groups
- here called Knowledge Nodes (KNs) -, and the coordin ation required in order
to exchange knowledge among them. Finally a peer to peer architecture to support
knowledge exchange across distributed and autonomous KNs is presented.
Keywords. Knowledge Management, Distributed Knowledge Management, Distributed and Federated Systems, Organizational Cognition, Peer to Peer.
1 Centralized versus Distributed Knowledge Management
I n c o m p l e x o r g a n i z a t i o n s , s u c h a s n e t w o r k e d a n d m u l t i -national firms - c o m p o s e d
by a large number of units, divisions and offices -, k n o w l e d g e m a n a g e m e n t ( K M ) i s
increasingly recognized as an important discipline to enable processes of creating,
c o d if y i n g a n d d i s s e m i n a t i n g k n o w l e d g e .
In the last ten years, many companies have invested a lot of resources and efforts
in KM projects, whose typical outcome is the creation of computer based systems
c o m p o s e d b y a K n o w l e d g e B a s e ( K B ) a n d a n E n t e r p r i s e K n o w l e d g e P o rtal (EKP) [6].
The aim of these projects is to create large, homogeneous reposito ries, in which corp o r a t e k n o w l e d g e i s m a d e e x p l i c i t , collected, represented, and organ i z e d , a c c o r d i n g
to a single, supposedly shared, system of meanings. This centralized representation o f t e n c a l l e d k n o w l e d g e m a p , o n t o l o g y , c a t e g o r i z a t i o n , o r c l a s s i f i c a t i o n s y s t e m - is
meant to represent a shared conceptu a l i z a t i o n o f c o r p o r a t e k n o w l e d g e , a n d t h u s t o
enable communication and knowledge exchange across the entire organization. From
a n e p i s t e mological perspective, these KM systems reflect an objectivistic view of the
w o r l d w h e r e m e a n i n g s a r e a s s u m e d t o b e u n i v ocal pictures of world objects. It pre supposes that all contextual, subjective, and social aspects of knowledge can be
elim in a t e d i n f a v o r o f a u n i q u e , o b j e c t i v e , a n d g e n e r a l c o d i f i c a t i o n .
Despite of this ap proach, a lot of theories of knowledge consider su b j e c t i v i t y a n d
sociality as intrinsic dimensions of knowledge (see [20], [9], [16], [15], [14], [25]).
They argue that knowledge is the result of different perspective and partial interpreta tions of world portions or domains, generated by individuals and groups through
so cial interactions. Therefore, it is proposed that knowledge, rather than being
viewed as an absolute monolithic matter, is better represented as a system of local
“ k n o w le d g e s ” c o n t i n u o u s l y n e g o t i a t e d b y c o m m u n i t i e s o f “ k n o w e r s ” [ 1 1 ] .
T h i s different epistemo l o g i c a l p a r a d i g m l e a d s t o a d i f f e r e n t a p p r o a c h t o K M c a l l e d D i s t r i b u t e d K n o w l e d g e M a n a g e m e n t ( D K M ) - i n w h i c h s u b j e c t i v i t y a n d so ciali t y a r e t a k e n a s i r r e d u c i b l e a s p e c t s o f k n o w l e d g e , a n d a r e v i e w e d a s a p o t e n t i a l
source of value, rather than as a problem to overcome [5], [7]. In DKM, an organiza tion is viewed as a “constellation” of local “knowledges” which reside within organizational communities, and live in the interplay between the need of sharing perspectives, in order to incrementa l l y i m p r o v e p e r f o r m a n c e s , a n d t h e n e e d o f m e e t i n g d ifferent perspectives, in order to sustain innovation [4]. Thus, the management of
k n o w l e d g e s h o u l d s u p p o r t a n d b a l a n c e t o v e r y g e n e r a l o r g a n i z a t i o n a l p r i n c iples:
− Principle of Autonomy: each community has a high degree of autonomy to manage its local knowledge. Autonomy can be allowed at different levels. We are
mainly interested in what we call semantic autonomy, namely the possibility of
c h o o s i n g t h e m o s t a p p r o p r i a t e c o n c e p t u a l i z a t i o n s o f w h a t i s locally k n o w n .
− Principle of Coordination: each community exchanges knowledge with other
units not by imposing the adoption of a single, common interpretative schema
(this would be a violation of the first principle), but by adopting a mechanism of
m e a n ing translation across different interpretative contexts, called semantic interoperation.
In this article we argue that, in designing KM systems, technological architectures
need to be consistent with the social process of organizational cognition (see section
2). M o r e o v e r , a s s u m i n g a d i s t r i b u t e d a p p r o a c h t o o r g a n i z a t i o n a l c o g n i t i o n , s u c h a
need is suggested as a major reason to explain the unsuccessful implementation of
current KM systems since based on a centralized view (see section 2.1).
In order two balance the needs of autonomy and coordination across different KNs,
n a m e l y o r g a n i z a t i o n a l e n tit i e s s u c h a s i n d i v i d u a l s a n d g r o u p s w h i c h o w n a “ l o c a l
knowledge”, a DKM architecture is proposed (see section 2.2). Finally, based on the
c o n c r e t e s e t t i n g o f a B a n k , a p e e r t o p e e r a r c h i t e c t u r e s u p p o r t i n g k n o w l e d g e ex change among autonomous KNs, is described (see section 3).
2 Technological and Social Architectures
As already underlined in [6] and in [24], technology and organizations are interrel a t e d d i m e n s i o n s s i n c e e a c h influences and structures the other. Here we underline
that the more an organizational process involves high level human activities, such as
o rg a n i z a t i o n a l c o g n i t i o n , t h e s t r o n g e r t h e i n t e r d e p e n d e n c y b e t w e e n t e c h n o l o g i c a l
a n d o r g a n i z a t i o n a l d i m e n s i o n s w i ll be. In particular, since each approach to cogni t i o n m a k e s s p e c i f i c ass u m p t i o n o n t h e r o l e o f c o m m u n i c a t i o n , a t e c h n o l o g y t h a t
strongly structures social communication implies a particular model on how cognition occurs [3]. More over, we agree with the structurationist approach (for more de tails see [21], [22]) which considers social and technological architectures as interd e p e n d e n t , r a t h e r t h a n i n d e p e n d e n t , d i m e n s i o n s w h i c h n e e d t o b e r e c i p r o c a l l y c o n s i stent.
In general, we argue that technological archite c t u r e s c a n b e d e s i g n e d a c c o r d i n g t o
two opposite approaches which assume two different social architectures of organiza tional cognition. The first one (centralized) views organizational cognition as a convergent process that collects peripheral "raw" knowledge, from various sources, and
codifies it into a central repository. The second one (distributed), as described by [4],
represents organizational cognition as a distributed process that balances the autonom o u s k n o w l e d g e m a n a g e m e n t o f i n d i v i d u a l a n d g r o u ps (perspective making), and
the coordination needed in order to exchange knowledge across different
autonomous entities (perspective taking). In the first case, technology is required to
e n a b l e c e n t r a l c o n t r o l , s t a n d a r d i z a t i o n , h i g h c a p a c i t y , a n d r o b u s t n e ss. In the second
one it should enable distributed control, differentiation, customization, and redundancy.
2.1 Centralized KM Architectures
Briefly we underline some of the main features that characterize a KM system
based on a cent ral i zed s oci al a rchit ecture of organizational cognition (for a more
d e tailed description see [6]). Socially, organizational cognition is viewed as a pro c ess whereby people:
− g e n e r a t e k n o w l e d g e t h r o u g h p e r i p h e r a l s o c i a l i z a t i o n i n c o m m u n i t i e s o f p r a c tices
[27], [11]: through work p r a c t i c e , e m p l o y e e s g e n e r a t e a n i m p l i c i t k n o w l e d g e i n
terms of working solutions that can be fruitfully made explicit and thus reusable;
− contribute with their knowledge through a codification process: knowledge is
cate g o r i z e d a n d v a l i d a t e d b y e x p e r t s a c c o r d i n g t o a c o r p o r a t e l a n g u a g e ;
− retrieve knowledge using a unified access to the organizational memory: through
the use of manuals, procedures, routines, or the access to formal t r a i n i n g , p e o p l e
have access to corporate knowledge.
F r o m a t e c h n o l o g i c a l p o i n t of view, architectures display some common features
t h a t a r e c o n s i s t e n t w i t h t h e a b o v e s o c i a l d e s c r i p t i o n o f o r g a n i z a t i o n a l c o g n ition (for
a more detailed overview on KM technologies see [13]). According to these organizat i o n a l a s p e c t s c e n t r a l i z e d K M a r c h itectures must:
− create and enable communication within formal and informal groups and commun ities. On-line interaction processes in terms of social/contextual cues [26] are e n abled by a large variety of tools. Indeed also through “virtual communities” and
g r o u p -w a r e a p p l i c a t i o n s i n d i v i d u a l s i n t e r a c t a m o n g e a c h o t h e r s a n d p r o d u c e t h e i r
“raw” peripheral knowledge;
− collect “raw” peripheral knowledge through workers participation. They can contribute to create and feed knowledge using automatic document management
tools, clustering, text mining, and information retrieval applications to explicit
a n d c o ll e c t k n o w l e d g e ;
− categorize and store knowledge in databases and repositories according to a comm o n a n d s h a r e d s y s t e m o f m e a n i n g , “ d i s t i l l i n g ” k n o w l e d g e t h a t i s u seful for the
whole organization;
− design a corporate system of meanings in the form of a common language, an o ntology, a knowledge map, a categorization, or a classification system that is necessary to codify knowledge according to a shared interpretative schema;
− create an Enterprise Knowledge Portal (EKP), that provides a unique, simple point
of access to corporate knowledge for the members of different organ izational units.
Most of the time, members access to the KB through various forms of personaliza tio n tools (e.g., individual or group profiles, views, chats, and so on).
As observed in a paradigmatic case study of a worldwide consulting firm d e scribed
in [5], these centralized KM systems are often deserted by end-users. It is shown that
a n y a p p r o a c h w h ich disregards the plurality of interpretative schemas is bound to
tro u ble, as the outcome will be perceived by users either as irrelevant (there is no
d e e p u n d e rstanding of the adopted and centralized schema) or as oppressive (there is
no agre e m e n t o n t h e unique schema, which is therefore rejected) [10].
2.2 Distributed KM Architectures
According to a distributed approach to KM, technologies should mainly support
the autonomous creation and organization of knowledge locally produced by individuals and gro ups and, on the other hand, support coordination processes among
autonomous entities, in order to exchange and share knowledge. In particular this
means:
− give to each knowledge owner the possibility to represent and organize knowle d g e a c c o r d i n g t o h e r g o a ls and interpretative perspective. Here, a DKM system
repre s e n t s e a c h k n o w l e d g e o w n e r a s a K n o w l e d g e N o d e ( p a r a g r a p h 2 . 2 . 1 ) ;
− provide tools to support the exchange of knowledge across different in dividuals
without assuming shared meanings but rather enabling the dynamic translation of
d i f f e r e n t m e a n i n g s . H e r e a D K M s y s t e m s u p p o r t s m e a n i n g n e g o t i a t i o n p r o c esses
(paragraph 2.2.2);
− s e t m e c h a n i s m s a n d p r o t o c o l s t o e n a b l e , t h r o u g h c o o p e r a t i o n , t h e e m e rg e n t a n d
b o t t o m -u p f o r m a t i o n o f i n f o r m a l c o m m u n i t i e s a n d c o m m u n i c a tion practices (such
a s f i n d i n g o r a d d r e s s i n g p e o p l e t o t r u s t e d i n d i v i d u als/communities). Here a DKM
system supports the formation of groups and knowledge discovery/propagation
t h r o u g h s o c i a l c o o p e r a t i o n ( p a ra graph 2.2.3).
In the following three para graphs we will describe these elements in more de tail
proposing some of the main requirements that should be considered in a DKM architecture (see Figure 1). In the next section, such requirements will be framed in the
s p e c i f i c c o n t e x t o f a k n o w l e d g e e x c h a n g e a p p l i c a t i o n d e v e lo p e d w i t h i n a B a n k .
Fig. 1. A DKM technological architecture
2.2.1 Knowledge Nodes
As we said above, a complex organization can be seen as a “constellation” of
g r o u p s a n d c o m m u n i t i e s , a n d , fr o m a d e s i g n i n g p o i n t o f v i e w , a s a “ c o n s t e l l a t i o n " o f
K N s c o n s i d e r e d i n [ 8 ] a s t h e b u i l d i n g b l o c k s o f a D K M a r c h i t e c ture.
A KN, has been described as the reification of organizational units - either fo rm a l
(e.g. divisions, market sectors) or informal (e.g . interest groups, communities of pra c tices, communities of knowing) - which exhibit some degree of semantic auto n omy.
Semantic auto n o m y m e a n s t h e a b i l i t y t o d e v e l o p a u t o n o m o u s i n t e r p r e t a t i v e s c h e m a s
(perspectives on the world) to interpret, organize, and store useful inform a tion. In
other words, each KN is a content interpreted within an interp r e t a t i v e c o n t e x t w h i c h
is the autonomous manifestation of the node's perspective. Interpretative con texts are
b o t h i n d i v i d u a l l y h e l d a n d s h a r e d , t h r o u g h n e g o t i a t i o n , b y g r o u p s a n d c o m munities.
Moreover, they become visible by looking at how individuals and groups use
tech n o lo g y i n o r d e r t o s t o r e , o r g a n i z e a n d u s e i n f o r m a t i o n . S o m e t i m e s g r o u p s a n d
c o m m u n ities develop shared repositories where content is stored and organ i z e d
t h r o u g h c o m m on t axonom i es (which are here considered as representations of the
c o m m u nity's interpretative schemas) while, some other, individual tech n o l o g i e s
(such as PCs) are considered as the primary artifact for knowledge organization. Thus,
w h e n c o n c r e t e l y l o o k i n g a t o r g a n izations (see the case of the bank after), existing
t e c h n o lo g y a n d l o c a l p a t t e r n s o f u s e , c a n s u g g e s t u s w h e t h e r t o c o n s i d e r a s r e l e v a n t
K N s i n d iv i d u a l s r a t h e r t h a n g r o u p s o r b o t h .
2.2.2 Meaning negotiation
I n a D K M a p p r o a c h , K N s are thus materialized by local technologies that represent a
semantically autonomous expression of local knowledge owned by an individual or
a group. In particular, within such local technologies, we observe, besides the sp e cific features provided to auto n o m o u s l y m a n a g e k n o w l e d g e ( s u c h a s v a l i d a t i o n a n d
c o n t ribution processes), the availability of a logical element that is relevant in order
to support coordination processes among KNs. Such an element, namely a context
[12], is a representation schema of th e s y s t e m o f m e a n i n g s a d o p t e d w i t h in each KN.
As proposed in [12], a context is defined as a meta-d a t a s c h e m a u s e d i n o r d e r t o i n terpret the local system of meanings, and make it syntactically comprehensible to
o t h e r s . F o r e x a m p l e , a c o n t e x t c o u l d b e t h e directory structure of a web site or the
taxonomy contained in an EKP. Contexts can be used by a DKM system in order to
enable knowledge exchange among KNs. In particular, through some technological
c o m p o nent which is able to extract, use and compare interp retative contexts, a KN
c a n re trieve information from other nodes even if stored according to a different se mantic schema [1], [23], [19]. Moreover, given an information request (query) from a
KN A, a software component should be able to contact KNs B and C and to compare
the meaning of the query within the context of A with the meaning that emerges
w i t h i n t h e c o n t e x t s o f B a n d C 1 . S u c h a p r o c e s s o f m e a n i n g a s s e s s m e n t a n d c o m p a ri s o n a c r o s s d i f f e r e n t c a t e g o r i z a t i o n s c h e m a s i s h e r e n a m e d m e a n i n g n e g o t i a t i o n p ro c ess (for an algorithm of meaning negotiation see [17]).
2.2.3 Cooperation
Given that KNs are semantically autonomous and able to communicate through
meaning negotiation, some other organizational capabilities are needed in order to
s u s t a i n t h e d i s t r i b u ted social process of knowledge creation. According to research
on informal communities [11] [27] [18], these organizational features provide a set of
social filters based on elements such as trust, membership or shared interests that are
necessary to avoid communication overflow while ensuring effectiveness. Here we
briefly propose some of these abilities:
− Federate: KNs should be able to spontaneously federate creating groupings and
communities of nodes that display a common interest. Such an interest could be
given by the goal of maximizing the opportunity to be found by other KNs (being
part of a visible group), by the need to certify the type or the quality of a knowledge (through the filtering of members), or the issue to protect content and secure
k n o w l edge access from unauthorized KNs (through access policies). K -Federations
can simplify interaction processes be c a u s e a r e q u e s t c a n b e s e n t t o a g r o u p r a t h e r
than to individual KN, decreasing the number of interactions, or because knowledge retrieved from a K N t h a t b e l o n g s t o a g r o u p h a s p r e s u m a b l y a c e r t i f i e d q u a l ity.
1
For example, as presented in [8], the query composed by the keyword sequence “antivirus
software” in the context of a directory structure available in A “/officeactivities/software/antivirus/antivirus-up-date/manuals” has a meaning which can be considered
as similar to a document containing the same keywords and stored under the directory of B
“/documents/security -system/antivirus/antivirus-up-date/manuals” and not to one stored in C
under “/products/soft-ware/anti-virus/marketreports/last” even if keywords are still compatible.
− D i s c o v e r : K N s s h o u l d b e a b l e t o f i n d e a c h o t h e r h a v i n g t h e o p p o r t u n i t y t o d is-
−
−
In
in
cover who is available, what type of services are accessible, and what kind of con ditions are necessary to establis h c o m m u n i c a t i o n . S u c h t y p e o f c a p a c i t y s h o u l d b e
distri b u t e d a c c o r d i n g t o t h e d i s t r i b u t e d n a t u r e o f o r g a n i z a t i o n a l k n o w ledge.
Propagate: a KN that looks for information, should be able not to directly interact
with all the KNs in order understand who has the required knowledge. Rather, she
should be able to trust that a limited number of interactions with neighbors give
sufficient chances to find the right counterpart. Such type of capacity becomes
p o ssible if KNs are able to propagate requests to other n odes or redirect requesting
KNs to potential targets.
Learn: a KN, that maintains semantic relations with other KNs, should be able to
remember such relations in order not to re -n e g o t i a t e t h e m e a n i n g o f a r e q u e s t .
S u c h a k n o w l e d g e c a n b e a l s o f r u i t f u l l y u sed to redirect requests to other nodes or
p r o p a gate their request to appropriate targets.
the following section we will propose how these requirements can be implemented
a concrete business case adopting peer to peer technologies.
3 Peer to Peer Architecture for a DKM System
In this section it is shown how the functional requirements of a DKM sy st e m d r i v e
the choice of a particular architectural pattern design (a peer to peer system) and an
underlying technology framework (the Jxta Project).
In particu l a r t h i s k n o w l e d g e e x c h a n g e s y s t e m i s u n d e r d e v e l o p m e n t w i t h i n t h e
b u s i n e s s s e t t i n g o f a n I t a l i a n N a t i o n a l B a n k 2 . For more details see [8].
In the DKM approach, a great emphasis is given to autonomy and coordination
a sp e c t s , s o t h a t e v e r y K N c a n m a n a g e h e r k n o w l e d g e , e x c h a n g e k n o w l e d g e t h r o u g h
meaning negotiation, and cooperate with other KNs in order to achieve her goals.
Compared to these aspects a peer to peer system is particularly suitable (see [2]) since
depicted with the following capabilities:
− supports autonomy: every member of the system is seen as a peer that manages
and has control over a set of local technologies, applications and services;
− is d y n a m i c : p e e r s a n d r e s o u r c e s c a n b e a d d e d o r r e m o v e d a t a n y t i m e ;
− is decentralized: the community of peers is able to achieve its goal independently
from any specific member or component;
− is cooperative: in order to join and use the system, every member must provide
re sources or services to the others.
F r o m a n i m p l e m e n t a t i o n p o i n t o f v i e w , w e f o c u s e d o n J X T A 3 , a set of open, generalized peer to peer (P2P) protocols that allow devices to communicate and collabor a t e t h r o u g h a c o n n e c t i n g n e t w o r k . T h i s P 2 P f r a m e w o r k p r o v i d e s a l s o a s e t o f p r o to c o l s a n d f u n c tionalities as a decentralized discovery system, a n a s y n c h r o n o u s p o i n tto -point messaging system, and a group membership protocol. A peer is a software
component that runs some or all the JXTA protocols; every peer has to agree upon a
2
This architecture is under development as part of EDAMOK, a joint project of the Institute
for Scientific and Technological Research (IRST, Trento) and of the University of Trento.
3
A P2P open source project started in 2001 and supported by Sun. http://www.jxta.org/
c o m mon set of rules to publish, share and access “resources" (like serv ices, data or
applications), and communicate among each other. Thus, a JXTA peer is used to
su p port higher level processes (based, for example, on organizational considerations)
that are built on top of the basic P2P network infrastructure; they may include the
e n hancement of basic JXTA protocols (e.g. discovery) as well as user-written applica t i o n s . J X T A t a c k l e s t h e s e r e q u i r e m e n t s w i t h a n u m b e r o f m e c h a n i s m s a n d p r o to cols:
for instance the publishing and discovery mechanisms, together with a messageb a s e d c o m m u n i c a tion infrastructure (called “pipe”) and peer monitoring services,
s u p p o r t s d e c e n t r a l i z a t i o n a n d d y n a m i s m . S e c u r i t y i s s u p p o r t e d b y a m e m bership
s e r v i c e ( w h i c h a u t h e n t i c a t e s a n y p e e r a p p l y i n g t o a p e e r g r o u p ) a n d a n a c c e s s p r o to col (for authoriza tion control).
These features of a P2P system, and in particular those provided by JXTA, seem to
match the spirit and the main functional aspects of a KN in a DKM application, su ggesting this architectural solu tion as a logical choice. In particular:
− K N s ca n b e s e e n a s p e e r s t h a t o w n a k n o w l e d g e w h i c h i s s t o r e d a n d o rg a n i z e d
through local technologies and applications;
− m e a n i n g n e g o t i a t i o n c a n b e v i e w e d a s a p e e r q u e r y r e s o l u t i o n s e r v i c e t h a t in volves the use of a local context in order to retrieve or provid e k n o w l e d g e t h r o u g h
intera c t i o n p r o to cols;
− cooperation can be implemented through the set of dynamic and heterogeneous
services that each peer provides to the others in order to, for example, support the
discovery of other peers.
In this section we propose a P2P architecture (called Kex: Knowle d g e E x c h a n g e
System) which is coherent with the vision of DKM. In Kex (i) each peer (called KPeer in KEx) provides all the services needed by a knowledge node to create and
o rg a n i z e i t s o w n l o c a l k n o w l e d g e ( a u t o n o m y ), and (ii) social structures and protocols
o f m e a n i n g n e g o t i a t i o n a r e d e f i n e d i n o r d e r t o a c h i e v e s e m a n t i c c o o r d i n a tion (e.g.,
when searching documents from other peers).
Fig. 2. The KEx system: interaction between Seeker and Provider roles.
3.1 K -P e e r R o l e s
I n K E x e v e r y K -Peer has the same main system’s functionalities that are provided
b y t w o m a i n r o l e s : t h e p r o v i d e r a n d t h e seeker role. Each K-Peer can act as a pro vider, as a seeker, or can play simultaneously both roles. Figure 2 shows the intera c t i o n b e t w e e n a K -Peer that plays the seeker role and some K -P e e r s t h a t p l a y t h e p ro vider role. The following sections explain in details the two roles and their intera c tion.
3.1.1 Seeker
T h e s e e k e r c o m p o n e n t a l l o w s t h e u s e r t o i n t e r a c t w i t h o t h e r K-P e e r s a n d KFederations (see 3.2.1) in order to search information. The seeker has access to a
c o n t e x t s r e p o s i t o r y ( 3 . 2 . 2 ) t h a t c o n t a i n s t h e l o c a l se mantic representations of the
u s e r ’ s k n o w l e d g e i n t h e f o r m o f t a x o n o m i e s , c a t e g o r y s t r u c t u r e s o r ontologies.
Through a context browser, the seeker supports the user in the formulation of the
q u e r y b y s e l e c ti ng t he cont ext (or part of it ) which is relevant in defining the mea n ing of the query. In particular a query is composed by a query ex pression and one
focus. A query expression is a list of one or more keywords entered by the user; the
q u e r y ex p r e s s i o n c o u l d b e e m p t y . T h e f o c u s i s a s e l e c t i o n o f a p o r t i o n o f a c o n t e x t
t h e u s e r h a s m a d e . T h e s e e k e r p r o v i d e s t h e u s e r w i t h a d i s c o v e r y m e c h a nism (see
3 .2.3) to find resources (K -Peers or K -Federations) that could be targeted in retrieving
i n f o r m a t i o n . T h e u s e r d e c i d e s t o s e n d t h e q u e r y t o s o m e o f t h e a v a i l a b l e K -Peers and
K-Federations (as showed in step 1 in figure 2). When she submits the query, the
seeker activates a sess i o n t h a t i s a s s o c i a t e d t o t h e q u e r y a n d i s i n c h a r g e t o r e s o l v e i t
using the meaning negotiation protocol (see 2.2.2). When a provider receives a query
i t p e r f o r m s t h e q u e r y r e s o l u t i o n t h r o u g h b o t h a s y n t a c t i c a n d a s e m a n t i c a l g o rithm
(see 3.1.2) in order to retrieve relevant documents (steep 2 in figure 2). The provider
sends back to the seeker the result set (step 3 in figure 2).
A provider could propagate the query received to other providers that are considered as “experts” about the query’s topic (step 4 in figure 2). The active seeker session can receive asynchronously several incoming replies from those providers that
have been selected/suggested by other peers, and collects results that are composed
by the aggregation of all those that have been received; each result is made up of a
list of document descriptors (name of the document, short description, and so on) and
the indication of the part of the providers’ context (focus) that has been used in order
to interpret the meaning of the query and provide a resolution. This relationship
b etween contexts can be used as a learning opportunity (see 3.2.5) that the seeker can
store and reuse for other queries. Finally the seeker allows the user to access to the K Peer download service; if the user finds in the result set one or more interesting
d o c u m e n t s , s h e c a n c o n t a c t t h e p r o v i d i n g K -P e e r t o d o w n l o a d i t .
3.1.2 Provider
The provider is the second main role in the KEx system. It contains all the features
required to accept and resolve a query identifying results that have to be sent back to
the seeker. When a K -P e e r r e c e i v e s a s e m a n t i c q u e r y ( c o m p o s e d b y k e y w o r d s a n d o n e
focus), it instantiates a provider configured to use a set of local contexts and docu m e n t s ( c o n t a i n e d i n a p a r t i c u l a r directory or repository) in order to resolve the query
in two ways:
• Semantic search: uses for each context a matching algorithm [17] that tries to find
a correlation between a provider’s context and the query focus. In particular the
m a t c h i n g a l g o r i t h m t ries to find in the provider’s context the focus that has a rele vant semantic relation with the one sent by the seeker. The focus that the provider
has used for the query resolution and related documents are returned as results. If
t h e f o c u s p o i n t s t o o t h e r K-peers, the provider will propagate the query.
• Syntactic search: uses an indexer (or any locally available keyword index) to
search for the occurrence of specific keywords into the set of documents owned by
the provider.
I f t h e s e m a n t i c q u e r y i s c o m p o se d o n l y f r o m k e y w o r d s , t h e p r o v i d e r w i l l u s e o n l y
the syntactic search, otherwise if it is composed only by a focus, the provider will use
only the semantic search. If both are available, the final result will be the merge of
semantic and syntactic results.
3.2 K -Peer Capabilities
B e s i d e s p l a y i n g t h e a b o v e r o l e s , a K -Peer has some capabilities such as create or
j o i n a K -F e d e r a t i o n , m a n a g e c o n t e x t s , a n d u s e a d i s t r i b u t e d d i s c o v e r y m e c h a n i s m , a
q u e r y p r o p a g a t i o n f u n c t i o n a l i t y a n d a l e a r n i n g a l g o rithm. All these capabilities will
be explained in the following paragraphs.
3.2.1 K -F e d e r a t i o n
A K-Federation is a group of K -P e e r s t h a t a g r e e t o b e c o n s i d e r e d a s a s o l e e n t i t y b y
o t h e r K-Peers when these are requesting services (such as semantic search) to the
former. In other words, a Seeker can send a query to a K-F e d e r a t i o n , a n d t h e q u e r y
will be forwarded to each K-F e d e r a t i o n m e m b e r . A s a c o n se q u e n c e , t h e r e s p o n s e o f
the K -Federation is the same as if the query was sent directly to all the members of the
K-Federation (even if in the returned result set is specified if the Provider answers as a
m e m b e r o f a K-Federation). The K-F e d e r a t i o n c a n b e t h o u g h a s a “ s o c i a l ” a g g r e g a t i o n o f K -P e e r s t h a t d i s p l a y s o m e s y n e r g y i n t e r m s o f c o n t e n t ( p r o v i d e s y n e r g i c c o n tent), quality (certified content) or access policies (certified members).
T o b e c o m e a m e m b e r o f a K-F e d e r a t i o n , a K-P e e r m u s t p r o v i d e a K-Federation
S e r v i c e ( f o r e x a m p l e t h e q u e r y r e s o l u t i o n s e r v i c e ) , i m p l e m e n t t h e r e q u i r e d KFederation's protocol (for example the commitment to reply to queries sent to the K F e d e r a t i o n ) a n d o b s e r v e s t h e K-Federation's membership policy (for example reply
only to queries coming from federation members).
3.2.2 Context Management
T h e c o n t e x t m a n a g e m e n t m o d u l e a l l o w s u s e r s t o m a n a g e c o n t e x ts in order to
search or provide information. Contexts provide a semantic classification of the
knowledge and data. At the system level, KEx avoids the need to create huge syntactic a l i n d e x e s a n d t o t r a n s f o r m k n o w l e d g e a n d i n f o r m a t i o n i n t o a n h o m o g e n e o u s
fo rm a t ; i t g i v e s t o e a c h K -P e e r t h e a u t o n o m y t o s t o r e a n d m a i n t a i n i t s o w n i n f o r m a tion in the way it prefers, providing a tool to enable autonomous semantic classification of con tent. Multiple views can be built over the local information, each corres p o n d i n g to different contextual representations and semantic classifications of con tent. Two components are provided to man age contexts:
1. Context editor: gives to the user the possibility to edit and manage a context.
Moreover when the user crates a context she c an classify information (documents,
a d dresses of other K-Peers) into the context. In particular the user creates the links
between a context's concept and the specific information stored in local repositories (or links to other peers or directly to external i n f o r m a t i o n m a i n t a i n e d i n o t h e r
K-Peers repositories). This activity (relating context to content) can be also auto m a t ic a l l y p e r f o r m e d b y a n y t e x t m i n i n g t o o l i f i n t e g r a t e d w i t h t h e p e e r s y s t e m
(e.g. able to transform the documents representation schema p r o v i d e d b y t h e m in ing e n g i n e i n t o a c o n t e x t u s i n g a s t a n d a r d s y n t a x s u c h a s C T X M L [ 1 2 ] ) .
2 . Context browser: allows the user to specify a query in order to search information
in the network. The user can select a context from a set of available contexts and
browse the structure of the selected one in order to focus on a particular con t e x t u a l
topic (focus). The focus represents a set of contextual information that the provid i n g K -Peers can use to interpret the meaning of a query (that can be expressed as a
t ra d i t i o n a l s e t o f k e y w o r d s ) a n d t o c o m p a r e i t w i t h t h e i r o w n c o n t e x t s .
3.2.3 Discovery
Discovery is a mechanism that allows the user to discover resources in the peer to
peer network. The user discovers K -Peers or K -F e d e r a t i o n s a v a i l a b l e i n t h e n e t w o r k i n
o r d e r t o c o n t a c t t h e m a n d s e n d t h e m q u e r i e s . A K-Peer advertises the existence of
resources publishing an XML document (advertisement). In the KEx system two type
resources are advertised:
• K-Peers that have a provider service to solve queries. The main elem e n t s o f t h e
advertisement are a description of a K-P e e r s c o n t e x t s a n d h o w t o c o n t a c t t h e K Peer in o rder to send a query or retrieve documents.
• K-Federations that are set of K-Peers that have a provider service to solve similar
q u e ries. The K -Federation a ssures that a query sent to a K -F e d e r a t i o n i s p r o p a g a t e d
t o a l l a c t i v e K-Peers that are member of the K-Federation. In this case the main
e l e m e n t s o f t h e a d v e rtisement are the K -F e d e r a t i o n t h e m e s , h o w t o c o n t a c t , h o w t o
join.
In order to discover resources in a peer to peer network, a K -Peer sends a discov ery
r e q u e s t t o a n o t h e r k n o w n K-Peer or sends a multi-c a s t r e q u e s t o n t h e n e t w o r k , a n d
receives responses (list of advertisements) that describe the available services and
re sources. It is possible to specify a searching criteria (such as a keyword or textual
e x p r e s s i o n ) t h a t i s m a t c h e d a g a i n s t t h e c o n t e n t s p r o v i d e d b y t h e a d v e r t i s e m e n t re lated to each K -Peer or K -F e d e r a t i o n d e s c r i p t i o n .
3.2.4 Query Propagation
T h i s f u n c t i o n a l i t y a l l o w s t h e K E x s y s t e m t o “ u se” trust and social relations as a mean
to find information in highly dynamic, heterogeneous and complex environment. In
fact a K-p e e r c a n t r u s t n o t j u s t t h e “ c o n t e n t ” k n o w l e d g e o f a n o t h e r p e e r , b u t a l s o i t s
“ r e l a t i o n a l ” k n o w l e d g e i n t e r m s o f t h e c a p a b i l i t y to redirect a request to other trusted
K-peers. Thus, when a Provider receives a query, it might propagate that query to
a n o t h e r P r o v i d e r i t c o n s i d e r s “ e x p e r t ” o n w h a t i t b e l i e v e s i s t h e m e a n i n g o f t h e re q u e s t . I n o r d e r t o d e c i d e w h e r e t o p r o p a g a t e a q u e ry a K -Peer has two possibilities:
− a “proximity” criteria: the query will be sent to known K -Peers (i. e. by using the
d isc o v e r y f u n c t i o n a l i t y ) a n d s e l e c t i o n w i l l b e d o n e a c c o r d i n g t o s o m e q u a n t i t a tive criteria (number of peers, number of possible re -r o u t i n g s – h o p s-, etc.); this
w a y K -Peers or providers that are non directly reachable by the Seeker or that have
j u s t j o i n e d t h e s y s t e m , c a n a d v e r t i s e t h e i r p r e s e n c e a n d c o n t r i b u t e t o t h e r e s o lu tion of the query;
− a semantic criteria: if the Provider com p u t e s s o m e m a t c h i n g b e t w e e n a q u e r y a n d
concepts in its own context, the query resolution mechanism might look for addresses of other K-P e e r s t h a t h a v e b e e n a s s o c i a t e d t o t h e m a t c h i n g c o n c e p t . H e r e
p r o p a g a t i o n i s d o n e o n t h e b a s e o f a n e x p l i c i t t r u s t s i n c e t h e provider defines
other peers as experts about the query topic.
The propagation algorithm combines the two possibilities and is based upon a
cost function that has the goal to ensure quality and reduce overflow. This goal is
a c h i e v e d p r i v i l e g i n g s e m a n t i c re -routing, and restricting the possibility to perform
proximity based hops the more semantic based hops are done.
Other parameters and mechanism controlling the scope of the search and prevent
t h e m e s s a g e “ f l o o d i n g ” a r e p r o v i d e d : t h e s e e k e r c a n s e t a t i m e to live (TTL), manu a l l y l i m i t t h e n u m b e r o f h o p s , s t o r e i n t h e q u e r y t h e n a m e o f K -Peers that have al ready received the query, and so on.
3.2.5 Learning
W h e n t h e m a t c h i n g a l g o r i t h m f i n d s a s e m a n t i c c o r r e s p o n d e n c e b e t w e e n c o n cepts
b e l o n g i n g t o d i f f e r e n t p roviders contexts, the Seeker can store this information for a
future reuse. This information is represented as a semantic “mapping” between con cepts, and can be used in three ways:
1 . w h e n t h e K -Peer receives a query from another seeker, it can reuse the corre s p o n d ing stored mapping to facilitate (or eventually don't perform) the matching alg o rithm;
2 . a P r o v i d e r c a n u s e t h e e x i s t i n g m a p p i n g t o f o r w a r d t h e q u e r y t o o t h e r K -Peers that
present a semantic relation with the topic of the query (see 3.2.4);
3 . in order to send the query to a trusted set of providers, the Seeker could search into
the mapping relations in order to select those that have been recognized as “experts” on a particular topic thanks to previous interactions. Such a system could be
viewed as a sort of semantic “bookmark” of preferred providers.
4 Conclusions and Research Issues
In this article we argued that technological architectures, when dealing with pro c esses deeply characterized by human communication, must be consistent with the
social architecture of the process itself. In particular, in the domain of KM, technol ogy must embody a principle of distribution that is intrinsic to the nature of organiza t i o n a l c o g n i t i o n .
As a consequence, we propose some requirements for designing and implementin g
technological architectures for KM, based on P2P technologies. Moreover, a number
of research issues emerge in order to map aspects of distributed cognition into tech n o logical requirements. Here we propose some of these:
− social discovery and propagation: in order to find knowledge, people need to
d isc o v e r w h o i s r e a c h a b l e a n d a v a i l a b l e t o a n s w e r t o a r e q u e s t . O n t h e o n e h a n d ,
broadcasting messages generates communication overflow, on the other talking
j u s t t o p h y s i c a l l y a v a i l a b l e n e i g h b o r s r e d u c e s t h e p o tential of a distributed net work. A third option could be for a seeker to ask his neighbors who they trust on a
topic and, among these, who is currently available. Here the question is about the
s o c i a l m e c h a n i s m s t h r o u g h w h i c h p e o p l e f i n d , o n t h e b a s e o f t ru s t a n d r e c o m m e n dation, other people to involve in a conversation. A similar approach could be
u s e d i n o rder to support the propagation of information requests. In this work we
propose an hypothesis of social/semantic discovery and propagation. Nonetheless
further re s e a r c h n e e d s t o b e d o n e i n o r d e r t o m a t c h t h e o r e t i c a l a n d p r a c t i c a l a n a l y sis of in formal in f o r m a t i o n f l o w s a n d t o m a p n e w m o d e l s o f b o t t o m -u p c o o r d i n a tion mechanisms into distributed technologies.
− b u i l d i n g c o m m u n i t i e s : i f w e c o n s i d e r c o m m u n i t i e s as network of people that, to
some extent, tend to share a common interpretative schema, mechanisms will be
n e e d e d t o s u p p o r t t h e b o t t o m -up emergence of semantic similarities across intera c ti n g K N s . T h r o u g h t h i s p r o c e s s , t h a t c o u l d b e b a s e d o n m e a n i n g n e g o t i a t i o n
p r o to c o l s , p e o p l e c o u l d d i s c o v e r a n d f o r m v i r t u a l c o m m u n i t i e s , a n d w i t h i n organizations, managers might monitor the evolving trajectories of informal cogni-
t i v e n e tw o r k s . T h a n s u c h n e t w o r k s c a n b e v i e w e d a s p o t e n t i a l n e i g h b o r h o o d s t o
su p port social d i s c o v e r y a n d p r o p a g a t i o n .
References
1. S. Bergamaschi, F. Guerra, and M. Vincini. A Data Integration Framework for e-Commerce
Product Classification P r o c e e d i n g s o f 1 s t I n t e r n a t i o n a l S e m a n t i c W e b C o n f e re n c e
(ISWC'2002), Ian Horrocks James Hendler Eds., Springer Press, 2002.
2. D. Bertolini, P. Busetta, A. Molani, M. Nori, A. Perini. Peer-to-peer, agents, and knowledge
management: a design framework I R S T T e c h n i c a l R e p o r t N o . 0 2 0 7 -15, Istituto Tren t i n o d i C u l t u r a , 2002.
3. R. J. Boland, R. V. Tenkasi, and D. Te'eni. Designing Information Technology to Support
Distributed Cognition, O r g a n i z a t i o n a l S c i e n c e , 5, 3 (August), 1994.
4. R. J. Boland, and R. V. Tenkasi Perspective Making and Perspective Taking in Communities
of Knowing, O r g a n i z a t i o n S c i e n c e , 6, 4 (July -August), 1995.
5. M. Bonifacio, P. Bouquet, and A. Manzardo. A distributed intelligence paradigm for know ledge management. In A A A I S p r i n g S y m p o s i u m S e r i e s 2 0 0 0 o n B r i n g i n g K n o w l e d g e
to Business Processes. AAAI, 2000.
6. M. Bonifacio, P. Bouquet, and P. Traverso. Enabling distributed knowledge management.
Managerial and technological implications. Informatik - Informatique, 1/2002.
7. M. Bonifacio, P. Bouquet, and R. Cuel. The Role of Classification(s) in Distributed Know ledge Management. To appear in the Proceedings of 6th International Conference on Know ledge-Based Intelligent Information Engineering Systems & Allied Technologies (KES'2002),
Special Session on Classification. September 16-18, 2002. Crema (Italy). IOS Press.
8. M. Bonifacio, P. Bouquet, and R. Cuel. KNs: the building blocks of a distributed approach to
K M J o u r n a l f o r U n i v e r s a l C o m p u t e r S c i e n c e , 2002.
9. P. Bouquet. Contesti e ra g i o n a m e n t o c o n t e s t u a l e . I l r u o l o d e l c o n t e s t o i n u n a
t e o r i a d e l l a r a p p r e s e n t a z i o n e d e l l a c o n o s c e n z a . Pantograph, Genova (Italy), 1998.
10. G. C. Bowker and S. L. Star. S o r t i n g t h i n g s o u t : c l a s s i f i c a t i o n a n d i t s c o n s e q u e n c e s .
MIT Press., 1999.
11. J. S. Brown and P. Duguid. Organizational learning and Communities of Practices: Toward a
Unified View of Working, Learning and Innovation. O r g a n i z a t i o n S c i e n c e , 2(1), 1991.
12. P. Bouquet, A. Donà, L. Serafini, and S. Zanobini Contextualized local ontologies specification via CTXML To appear in the working notes of the AAAI-02 workshop on Meaning
Negotiation. Edmonton (Canada). July 28, 2002.
13. U.M. Borghoff and R. Pareschi I n f o r m a t i o n T e c h n o l o g y f o r K M Springer, Berlin, 1999.
14. J. Dinsmore. P a r t i t i o n e d R e p r e s e n t a t i o n s . Kluwer Academic Publishers, 1991.
15. G. Fauconnier. Conceptual Integration Networks. C o g n i t i v e S c i e n c e , 22(2):133-187,
1998.
16. C. Ghidini and F. Giunchiglia. Local Models Semantics, or Contextual Reasoning = Locality
+ Compatibility. Artificial Intelligence, 127(2):221-259, April 2001.
17. B. Magnini, L. Serafini, and M. Speranza. Linguistic based matching of local ontologies. To
appear in the working notes of the A A A I-0 2 w o r k s h o p o n M e a n i n g N e g o t i a t i o n .
Edmonton (Canada). July 28, 2002
18. J. Lave and E. Wenger. S i t u a t e d L e a r n i n g : L e g i t i m a t e P e r i p h e r a l P a r t i c i p a t i o n .
New York, Cambridge University Press. 1990.
19. J. Madhavan, P. Bernstein, P. Domingos, and A. Haley, Representing and Reasoning about
Mappings between Domain Models at the E i g h t e e n t h N a t i o n a l C o n f e r e n c e o n A r t i f i cial In telligence (AAAI'2002), Edmonton, Canada.
20. J. McCarthy. Notes on Formalizing Context. In P r o c . o f t h e 1 3 t h I n t e r n a t i o n a l J o i n t
C o n ference on Artificial Intelligence, pages 555-560, Chambery, France, 1993.
21. W.J. Orlikowski and D.C. Gash. Technological Frames: Making Sense of Information Technology in Organization. A C M T r a n s a c t i o n s o n I n f o r m a t i o n S y s t e m s , 1994 Vol.12/2,
April, pages 174-207.
22. W. J. Orlikowsy, and D. Robey Information Technology and the Structuring of Organizations I n f o r m a t i o n S y s t e m s R e s e a r c h , 2, 2 (June), 1991.
23. S. Prasad, Y. Peng, and T. Finin. A Tool For Mapping Between Two Ontologies Using
Explicit Information. Notes of the A A A I-0 2 W o r k s h o p o n M e a n i n g N e g o t i a t i o n , Edmonton, Canada, August 2002.
24. R.L. Purvis, V. Sambamurthy, and R.W. Zmud. The Assimilation of Knowledge Platforms
in Organizations: An Empirical Investigation. O r g a n i z a t i o n S c i e n c e , 2001, Vol 12, No 2,
March-April, pages 117-135.
25. W.V. Quine. Propositions. In O n t o l o g i c a l R e l a t i v i t y a n d o t h e r E s s a y s . Columbia
University Press, New York, 1969.
26. L. Sproull and S. Kiesler. Reducing Social Context Cues: Electronic Mail in Organizational
Communication. Irene Greif (Eds.), C o m p u t e r S u p p o r t e d C o o p e r a t i v e W o r k : A B o o k
o f R e a d i n g s, San Mateo CA, Morgan Kaufman Publishers, 1988.
27. E. Wenger. C o m m u n i t i e s o f P r a c t i c e . L e a r n i n g , M e a n i n g a n d I d e n t i t y , Cambridge
University Press, 1998.