Subjectivity and Objectivity in Urban Knowledge Representation

Antonia Cataldo

, Valerio Cutini

, Valerio Di Pinto

and Antonio M. Rinaldi

4,5

Ministero dell’Istruzione, dell’Universit´a e della Ricerca, Via Ponte della Maddalena 55, 80142 Napoli, Italy

Dipartimento di Ingegneria Civile, Architettura e Urbanistica, Universit´a di Pisa, Via Diotisalvi 2, 56122 Pisa, Italy

Dipartimento di Ingegneria Civile, Edile e Ambientale,

Universit´a di Napoli Federico II, Via Claudio 21, 80125 Napoli, Italy

Dipartimento di Ingegneria Elettrica e delle Tecnologie dell’Informazione, Via Claudio 21, 80125 Napoli, Italy

IKNOS-LAB Intelligent and Knowledge Systems,

LUPT, Universit´a di Napoli Federico II, Via Toledo 402, 80134, Napoli, Italy

Keywords:

Conﬁgurational Analysis, Concept Mining, Knowledge Engineering, Ontologies, Space Syntax.

Abstract:

The question of subjectivity and objectivity of information is an important open issue in the knowledge engi-

neering research community. In the context of space representation, they have been traditionally considered

competing themes in the study of places, particularly in urban ones. This is highlighted by the distance, in

terms of cultural training and operational approach, between the professionals of the city: urban planners

and urban anthropologists. The growth in modeling capabilities allows a quantitative study of a city but in-

formation about the meanings of space elements are often not taken into account. Starting from this basic

assumption, our paper aim is to give a novel point of view to integrate subjectivity and objectivity in an oper-

ational model. Space Syntax, as a theory and a methodology, is used as a tool to study the objectivity of the

urban space. Ontologies, as an approach and a method to formally represent knowledge, is used to provide

Space Syntax with the subjectivity of the same spaces.

1 INTRODUCTION

In the era of big data, the use of formal models and

techniques is a necessary task to represent and man-

age large amount of information. In the context of

spatial information this is a hard issue due to the con-

tinuum nature of this kind of data. Our interest is fo-

cused on cities, as complex entities. In fact, they seem

to challenge descriptions, so that they can hardly be

deﬁned and treated in disciplinarily terms. Planners

and urban designers have always tried to use simpli-

ﬁed concepts and notions, thus emphasizing hierar-

chies, regular geometries and the separation of parts

from wholes. In this broad framework, one approach

that is increasingly engaging scholars and practition-

ers rethinks the way we can look at the city and its

problems and potential; another approach that starts

from space and urban life at the same time, hence

grasping the two dimensions of a place: cognitive and

geographic. The city is considered as an uniform fact.

For this purpose we need to imagine cities as com-

plex system consisting of many variables that inter-

act with each other, assuming the space as a primary

element in such dynamics rather than the mere and

inactive background of social and economic phenom-

ena. Conﬁgurational approach provides a concise and

effective overview on how a city operates, but it is

unable to render a formal representation of the mean-

ing of urban elements In the recent years, several ap-

proaches have been proposed to represent knowledge.

Some of them, based on ontologies, aim at deleting, or

at least smoothing conceptual or terminological mess

and actually provide a common view of the same in-

formation. The ontological aspects of information are

intrinsically independent from information represen-

tation, so that information itself may be isolated, re-

covered, organized and integrated with respect to its

contents (Rinaldi, 2008). The traditional approaches

to conﬁgurational analysis are based on the idea that

it is how things are put together that matters (Hillier,

B., 1996). From this point of view, the semantic of

built environment is completely neglected.

In this paper we introduce a new informative layer

based on a formal description of the elements spread

throughout the city. We propose a conceptual posi-

tion, and its capability to be used in an operationalized

411

Cataldo A., Cutini V., Di Pinto V. and M. Rinaldi A..

Subjectivity and Objectivity in Urban Knowledge Representation.

DOI: 10.5220/0005152404110417

In Proceedings of the International Conference on Knowledge Discovery and Information Retrieval (KDIR-2014), pages 411-417

ISBN: 978-989-758-048-2

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

model in order to integrate space syntax and space se-

mantic.

2 THEORY AND TOOL TO CITY

INVESTIGATION

Conﬁgurational analysis was pioneered as a general

attempt to relate urban physical dynamics and social

process. Its genesis was in the 1960’s urban revo-

lution, as a way to dominate the contradiction be-

tween the impressive architectures that were ﬂourish-

ing and the un-urban nature of their spaces. Among

different theoretical speciﬁcations and operative tech-

niques, the most important conﬁgurational approach

is known as Space Syntax. As a theory and methodol-

ogy, it is widely applied and appreciated for its capa-

bility of extracting cognitive data and information out

of the spatial layout of urban settlements.

Space Syntax indicates an objective way to inves-

tigate and deﬁne the relationships between the phys-

ical structure of man-made environments and social

structures or phenomena (Hillier, B. and Hanson, J.,

1984). Space syntax theorises that certain conﬁgura-

tional measures of centrality in graphs express a po-

tential to embody or transmit social ideas, and then

pushes this potential to spatial structures, by geo-

graphically linking the graph to the space. Many

years of research have highlighted that centrality is

related to several aspects in human interaction (Free-

man, 1978). There are three main distinct intuitive

conceptions of point centrality, and many measures

for each of them. The simplest and perhaps the most

intuitively conception is that centrality is function of

the degree of a point on the net, that is the number

of connections a node has to other nodes. It is gen-

erally known as Degree Centrality. Another simple

idea is that the independence of a point is determined

by its closeness to all other points in the graph: the

Closeness Centrality. Another view of point central-

ity is based upon the frequency with which a point

falls between pairs of other points on the topologi-

cal geodesic paths connecting them, in the idea that

it has a potential to control their communication: the

Betweenness Centrality. Space Syntax is mainly piv-

oted on closeness centrality, in terms of Integration

Index. It shows the cognitive complexity of reach-

ing a street, and is often argued to predict the pedes-

trian use of a street: the easier it is to reach a street,

the more popular it should be. Very interesting and

distinctive is also the way Space Syntax performs the

network making process. It is based upon the reduc-

tion of the city to the set of maximum size convex

spaces in which the urban public space is divisible.

On this map, known as the convex map, the minimal

set of lines crossing the convex spaces that oversee

them all is drawn, getting the axial map. It is an in-

version of the urban graph: axial lines represent the

nodes of the graph, while their intersections are the

edges. On this net centrality measures could be per-

formed. To associate their meaning to the real urban

space, it is therefore necessary to take the opposite

road. This method is known as dual or indirect ap-

proach. Several different operational techniques have

been so far developed, all sharing the base assump-

tion that it is the urban space, according to the way it

results from the arrangement and alignment of blocks

and buildings, what primarily involves the precondi-

tion for its use, thus strongly inﬂuencing the whole

inner geography of the settlement. On this basis, such

approach allows drawing out of the grid conﬁguration

speciﬁc parameters, which were proved suitable for

reproducing a wide range of urban aspects. Besides

its effectiveness as a knowledge extraction tool, this

capability also makes conﬁgurational analysis suit-

able for supporting town planning, in that it allows

simulating and predicting the effects of any planned

transformation of the physical consistency of the grid.

Since it appears confronting the traditional territorial

modelling, based on a strictly interactional approach,

it is no wonder that the introduction of a conﬁgura-

tional point of view was received as somehow hereti-

cal and likely to raise discords and criticisms, which

havenourished strong debate and lengthy discussions.

Far from increasing such debate, the explanatory

limits and the practical utility of such a successful

(and widely misinterpreted) approach appear worth

highlighting. To this aim, we will try to deﬁne it by

subtracting; namely what it’s not. We daresay that

Space Syntax is not a whole theory of the city in it-

self. Such a theory, in fact, should be able to grasp

the essence of city as a whole, thus encompassing

any functional and aesthetic aspect, so as to account

for forms and reasons of urban genesis and develop-

ment; what space syntax, actually a “theory of soci-

ety and space” (Hillier and Netto, 2001), cannot really

achieve. Moreover, Space Syntax does require some

speciﬁc theoretical position on city’s status to be ex-

plained and properly used: the expression of a set of

social relations forming part of a wider phenomenon

or structure. This assumption de facto creates the ex-

tent of Space Syntax boundary, re-deﬁning the con-

cept of space, the way it is inserted in the process

of urban transformation and the way the latter is ac-

counted: through a non-discursive approach. Never-

theless, space syntax is not even a mathematical the-

ory of the urban space. It uses mathematic (and its

graph theory elements) to explore the physical space

KDIR2014-InternationalConferenceonKnowledgeDiscoveryandInformationRetrieval

412

due to the admission that we are “extremely good at

using relational systems [...], but rather bad at know-

ing how to talk about them” (Hillier, B. and Hanson,

J., 1984). Much more properly than a theory in it-

self, space syntax can so be regarded as a theoretical

approach and a robust operational technique based on

the assumption that the grid of urban paths is the ac-

tive mechanism where social relations occur, suitable

for investigating and supporting the understanding of

the social logic of the city as a cultural product.

On such basis, several different operational tech-

niques have been so far introduced, so as to shape the

conﬁgurational approach as a widely ramiﬁed tree.

In order to contribute to the development of this ap-

proach, three directions are actually feasible. The ﬁrst

is to concur in increasing the height of the tree, reﬁn-

ing the operational tools and making them more pow-

erful and friendly. The second is to widen the crown

of the tree so as to eneble it cover and loom up more

and more territorial issues and increase its diffusion.

The third is to try strengthening its roots by working

on conceptual issues, aiming at clarifying and then

expanding the epistemological and operative bound-

ary of the discipline. This latter purpose will be dis-

cussed, in the conviction that the future sustainability

of the approach could beneﬁt from the overcoming of

the logic of wonderful isolation that has gone charac-

terizing space syntax, mainly due to the fuzziness of

its foundations.

The main aim of this paper is to explore the pos-

sibility of integrating space syntax with tools capa-

ble of returning the semantic aspects of urban space;

with the idea that a heuristic model, as space syn-

tax is, would greatly beneﬁt from the direct access

to the semantic information asset of the same space it

does syntactically analyze. The access to the seman-

tic information layer is, however, complex in itself in

that it needs to seize them by means a generalization

process of the locally acquired knowledge (through

the individual spatial experience). Our idea is based

on two essential cornerstones: the role of local com-

munities in shaping, and then meaning ascribing, the

urban space, and the representation objectivity such

meanings must be provided with. The ﬁrst one is a

conceptual issue ascribable to the scope of townscape.

As a fundamental concept in our proposition, it will

be discussed in the following paragraph. The second

one concerns the way we intend to acquire any seman-

tic information. This will be done to turn the concept

of ontology.

3 THE ROLE OF SUBJECTIVITY

IN TOWNSCAPE GENERATION

In this section we stress the notion of urban land-

scape known as perceived space. We highlight the

role of community in urban making and the neces-

sity of describing this cultural process by means of

a set of measures on city-networks (Space Syntax).

“Landscape means an area, as perceived by people,

[...]”(Europe, 2000): the European Landscape Con-

vention deﬁnes landscape stressing the basic role of

community perception of the surrounding environ-

ment. Moreover, many recent studies try to deﬁne

the quality of landscape paying speciﬁc attention to

the role it plays in the whole well-being of people,

as related with physiological and cognitive elements

(Velarde et al., 2007).

When we consider urban landscape, the question

of perception becomes highly complex due to the sev-

eral involved components, which arouse different per-

ceptive reactions. Following the Bourassa’s schema

(1990), the types of response to external perception

can be summarized into three categories: instinctive,

emotional and intellectual. This issue suggests that

when it comes to investigating on the concept of per-

ception, different types of reaction arise in the com-

munity related to aesthetics, emotional components

and cultural factors. Humans tend to prefer land-

scapes whose characteristics are easily interpretable.

The visual preferences are the result of emotional per-

ception given by the ease of retrieving information

about the surrounding environment.

From this prospective, the costruction of a for-

mal knowledge structure to represent landscape ele-

ments is a mandatory task. In the process of land-

scape understanding, we can identify two main tasks

at a cognitive level: assigning meanings to landscape

elements (making sense) to predict (no surprise) what

might happen (Kaplan, 1995); feeling attracted and

involved (involvement) to feel a sense of challenge.

The community has a crucial role in the way we relate

to landscape. People try to give a meaning to every el-

ement to easily structure the knowledge of what is vi-

sually perceived. At the same time, people feel a great

sense of attraction due to the complexity derivedfrom

the variety and heterogeneity of the elements that are

part of the territory. Referring to the theory of Ul-

rich (1984), the visual preferences follow a parabolic

trend: they increase as the number of elements, but

until the complexity does not reach too high values.

Moreover, when people go into the space, they

have a 3-dimensional perception of it: the capabil-

ity to give a meaning is provided by the readable-

ness while the feeling of being attracted shall be gen-

SubjectivityandObjectivityinUrbanKnowledgeRepresentation

413

erated by the mystery. A scene could be accounted

as readable if it has a homogeneous allocation of el-

ements that could be mapped (landmarks) as inter-

preted. In which case, the scene can be interiorised

and explored. The mystery is driven both by the ex-

ploratory sense (novelty) and by what we sense might

be seen. We can so argue that visual perception has

a key-role in this process. As already pointed out,

the community perception of the urban environment

stems not only from reading (more or less easily) the

elements it is composed of, but induces a mechanism

of affective interpretation. The affective perception

descends from processes of education, socialization

and acculturation, which deeply inﬂuence human be-

havior. Perception is screened by personal experi-

ence, age, culture, and so on, leading to non-unique

answers on the same visual stimuli. The affective per-

ception combines to bring about the sense of belong-

ing to a territory. Landscape is relished if it is com-

posed of coherent and effective elements (Coeterier,

1996): the being of a singularity (e.g. a historic build-

ing) is not enough to make a landscape charming, its

harmony within the neighbourhood is the key.

The knowledge everyone gathers in life further-

more inﬂuences the relationship between human and

environment, and the landscape perception in itself.

The landscape components assume a local sense due

to their relation to the human educational process.

Landscape has become an essential tool to the charac-

terization of community members due to its cultural

leading. The human perception of the anthropic en-

vironment (signally the urban one) arises as a rela-

tion of a large set of different reading and interpreta-

tion forms strictly tied to the hosting community and

the cultural changes in time. The mutation in urban

environment could be seen as a key-process into the

identity afﬁrmation of social groups. A city, as an

anthropic landscape, has elements in itself contribut-

ing to support the social and cultural cohesion of the

community that made them.

4 A CONCEPTUAL APPROACH

TO CITY REPRESENTATION

Once it is acknowledged that the process of under-

standing the space is a priority and a prerequisite

for the determination of actions for its management,

we need to identify the homogeneous territorial con-

texts that contain highly related and characterizing

factors. Therefore, the process of knowledge acqui-

sition about a territory starts from the recognition of

its elements and from their interpretation depending

on their context.

Starting form these considerations, we must try

to resolve conceptual misunderstandings and seman-

tic ambiguities and also generate a precise and accu-

rate description of our knowledge. In this context, the

deﬁnition of a common and shared ”language” is the

ﬁrst step in a knowledge formalization about the ur-

ban space. From our point of view, ontologies are

an efﬁcient and effective tool to address this issue.

In the last years ontologies have been studied in the

context of spatial knowledge and they have been used

in different domains to represent spatial properties,

data interoperability or relations among different data

sources (Bateman and Farrar, 2004; Janowicz et al.,

2013). In the authors’ opinion, any landscape has a

cultural meaning since it represents the main product

of a community in time. We use ontologies to rep-

resent this kind of knowledge shared in a community

and to give a formalization of all the elements spread

throughout the city.

Our approach starts from the modeling view of

knowledge acquisition (Clancey, 1993), where the

modeling activity must establish a correspondencebe-

tween a knowledge base and two separate subsys-

tems: the agent’s behavior (i.e., the problem-solving

expertise) and its own environment (the problem do-

main). This vision is in contrast with the transfer

view, wherein a knowledge base is a repository of

knowledge extracted from one expert’s mind. Us-

ing the modeling view approach, knowledge is much

more related to the classical notion of truth as corre-

spondence to the real world, and it is less dependent

on the particular way an intelligent agent pursues its

goals. Although knowledge representation is a basic

step in the whole process of knowledge engineering,

a part of the AI research community seems to have

been much more interested in the nature of reason-

ing than in the nature of “real world” representation.

The dichotomy between reasoning and representation

is comparable with the philosophical distinction be-

tween epistemology and ontology, and this distinc-

tion is important to better understand our aim and ap-

proach. Epistemology can be deﬁned as “the ﬁeld of

philosophywhich deals with the nature and sources of

knowledge” (Nutter, 1998). The usual logicistic inter-

pretation is that knowledge consists of propositions

whose formal structure is the source of new knowl-

edge. The inferential aspect seems to be essential to

epistemology: the study of the “nature” of knowledge

is limited to its superﬁcial meaning (i.e., the form),

since it is mainly motivated by the study of the in-

ference process. Ontology, on the other hand, can be

seen as the study of the organization and the nature of

the world independent of the form of our knowledge

about it.

KDIR2014-InternationalConferenceonKnowledgeDiscoveryandInformationRetrieval

414

A formal deﬁnition of ontology as highlighted in

(Gruber, 1993) is “a formal and explicit speciﬁca-

tion of a shared conceptualization”; conceptualiza-

tion refers to an abstract model of a speciﬁc reality

in which the component concepts are identiﬁed; ex-

plicit means that the type of the used concepts and the

constraints on them are well deﬁned; formal refers to

the ontology propriety of being “machine-readable”;

shared refers to the fact that an ontology captures the

consensual knowledge, accepted by a group of per-

sons. We also consider other deﬁnitions of ontology

(Neches et al., 1991). This deﬁnition indicates the

way to proceed in order to construct an ontology: i)

identiﬁcation of the basic terms and their relations; ii)

agreeing on the rules to arrange them; iii) deﬁnition of

terms and relations between concepts. From this per-

spective, an ontology includes not only the terms that

are explicitly deﬁned in it, but also those one that can

be derived using deﬁned rules and properties. Thus

an ontology can be seen as a set of “terms” and “re-

lations” among them, denoting the concepts that are

used in a speciﬁc domain.

In the context of spatial information, we should

stress that urban reading and interpretation lead to

three basic questions. The ﬁrst question refers to the

territorial context: the discussion of city has an on-

tological nature and refers to several signiﬁers. The

second question has a semiotic nature: the discussion

of city is based on the representation of a territorial

context (the real world) that consists of signs. The last

and surely most theoretical question is epistemologi-

cal and refers to meanings (concepts). From a strictly

operational view, these three questions reﬂect many

procedures. All general territorial sciences start with

the generation of a model of reality to evaluate the

meaning given to things, then conclude with an in-

tervention in the reality itself. Signs, meanings, and

signiﬁers are the triad upon which the representation

sciences are based, and these ideas are tightly con-

nected with each other. A sign produces models of

intervention; targets and values followed by a planner

affect and boost the building of meanings; meanings

determine the criteria and modalities of intervention,

which will affect reality and will change it; chang-

ing the reality changes the meanings; different mean-

ings imply different signs, which produce new rep-

resentations of reality. Thus, to interpret or deﬁne a

landscape is to translate and simplify the complex-

ity of space into decoded signs, meanings, and sig-

niﬁers. Giving a meaning to an object (signiﬁer) is

not an easy action. The meanings include two con-

cepts suggested in (Eco, 1968): a denotative element

and a connotativeelement. To denote an object means

to infer the function (meaning) of said object (signi-

ﬁer); we have an immediate communication because

the denotative meaning does not lead to ideologies or

meta-discussions. In contrast, the meaning has a con-

notative function when it expresses an ideology in a

potentially implicit or hidden way. It is the mean-

ings that refers to symbols, values, cultural products,

and intangible culture. Finally, we can assess the idea

that through forms it is possible to recognize the story

of objects, the things that remain from past societies.

Any object of a city, once recognized, gains the status

of a sign, and so the need to be interpreted. This in-

terpretation should not be limited only to recognizing

single elements (through decomposingpraxis), but in-

stead should refer to the context to which the signs

belong (relationship with the whole) or the ways in

which they have meaning and functionality.

Moreover, we are are interested also in the for-

malization of the concept of relevance information.

We can divide relevance into two main classes (Har-

ter, 1992; Swanson, 1986) called objective (system-

based) and subjective (human (user)-based) rele-

vance, respectively. Objective relevance can be

viewed as a topicality measure, i.e. a direct match

of the topic of the retrieved document and the one

deﬁned by the query. Several studies on human rel-

evance show that many other criteria are involved in

the evaluation of relevance (Barry, 1998; Park, 1993;

Vakkari and Hakala, 2000). In particular subjective

relevance refers to the intellectual interpretations car-

ried out by users and it is related to the concepts of

aboutness and appropriateness of retrieved informa-

tion. In addition, according to (Saracevic, 1996), ﬁve

types of relevance exist: an algorithmic relevance be-

tween the query and the set of retrieved information

objects; a topicality-like type, associated with the con-

cept of aboutness; cognitive relevance, related to the

user information need; situational relevance, depend-

ing on the task interpretation; and motivational and

affective relevance, which is goal-oriented. The dif-

ferent aspects of relevance can help in the deﬁnition

of a whole characterization of urban environment.

All these considerations can be summarized using

a formalization of objective knowledge expressed by

spatial syntax and a subjective knowledge represented

with ontologies. The ontology concepts arise from the

knowledge shared in a community. Considering the

conﬁgurational indexes as previously described, and

leaning them to a speciﬁc geographic location (i.e. a

location on the network), the same will take on a local

signiﬁcance. In such a case, the term “local” means

“in relation to a speciﬁc context”. Otherwise, in con-

ﬁgurational analysis, the same term “local” refers to

a method of index calculation. Speciﬁcally, it means

to reduce the analysis radius to a limited number of

SubjectivityandObjectivityinUrbanKnowledgeRepresentation

415

topological steps or to a predetermined metric neigh-

borhood, in order to explore the city at different scale

levels (Hillier, 2009). On this assumption, consider-

ing the existence of a local urban element, formally

described by an ontological model (e.g. (Cataldo and

Rinaldi, 2010)), we can deﬁne the Conﬁgurational

Ontology as:

CO = O(C

) (1)

Where O is the ontology as represented in our model,

and C

are the following network point centrality

measures (Freeman, 1978):

∑

i=1

a(p

, p

); C

n− 1

∑

i=1

d(p

, p

)

; C

∑

i=1

∑

j>i

(

)

where:

a(p

, p

) =

(

1, if p

andp

are connected by a line

0, otherwise

d(p

, p

) = the number of edges in the geodesic linkingp

and p

= the number of geodesics linking p

andp

) = the number of geodesics linking p

andp

that contain p

The above conﬁgurational indexesC

are the topolog-

ical attributes of the novel Conﬁgurational Ontology.

Unlike the others, our approach uses a combina-

tion of centrality measures quantitatively inﬂuenced

by the meaning of urban elements, understood as

“events” on the city network. In our ontology of the

city, in fact, each of the above events has a numeri-

cal attribute directly related to the people ﬂows which

generates its own. This allow us to make the urban

elements an endogenous part of the city model.

5 CONCLUSION

The proposed paper has investigated some issues re-

lated to the understanding of urban space following an

ontology-based approach. This process, like all those

based on cognitive mechanisms, is strongly affected

by the local nature of knowledge: the subjectivity.

Studies conducted in cognitive ﬁeld return a complex

fuzzy frame related to the difﬁculty of dealing with

the issue of subjectivity in a quantitative way. On the

other hand, studies dealing with the space in a purely

quantitative way, proposing to derive the subjectiv-

ity through the post-processing of numerical data, are

promising, but present several shortfall in some previ-

ous highlighted points. In our vision a possible solu-

tion is based on the concept of ontology used to give

a formal and common structure to express the seman-

tic complexity of the space. We consider both ob-

jective aspects of urban space using an algorithmic

approach and subjective issues related to the percep-

tion of communities which change the city, recurring

to the formalism of ontology. We give an integrated

representation of urban elements both from a concep-

tual and topological point of view using ontologies

and a conﬁgurational analysis tool (i.e. Space Syn-

tax). Combining these information we propose a pre-

liminary formal description of knowledge about a city

by means of a Conﬁgurational Ontology.

REFERENCES

Barry, C. L. (1998). Document representations and clues to

document relevance. Journal of American Society for

Information Science, 49(14):1293–1303.

Bateman, J. and Farrar, S. (2004). Towards a generic foun-

dation for spatial ontology. In Proc. of the 3rd Int.

Conf. FOIS’04, pages 237–248.

Cataldo, A. and Rinaldi, A. M. (2010). An ontological ap-

proach to represent knowledge in territorial planning

science. CEUS, 34(2):117 – 132.

Clancey, W. (1993). The Knowledge Level Reinterpreted:

Modelling Socio-Technical Systems. IJIS, 8:33–49.

Coeterier, J. (1996). Dominant attributes in the perception

and evaluation of the dutch landscape. Landscape and

urban planning, 34(1):27–44.

Eco, U. (1968). La struttura assente. La ricerca semiotica

e il metodo strumentale. Bompiani, Milano, Italy.

Europe, C. O. (2000). European landscape convention. In

Report and Convention.

Freeman, L. (1978). Centrality in social networks concep-

tual clariﬁcation. Social Networks, (1):215–239.

Gruber, T. R. (1993). A translation approach to portable on-

tology speciﬁcations. Knowl. Acquis., 5(2):199–220.

Harter, S. P. (1992). Psychological relevance and informa-

tion science. Journal of American Society for Infor-

mation Science, 43(9):602–615.

Hillier, B. (2009). Spatial sustainability in cities. organic

patterns and sustainable forms. In Proc. of SSS’09,

pages 1–20, Stockholm, Sweden. KTH.

Hillier, B. and Netto, V. (2001). Society seen through the

prism of space: outline of a theory of society and

space. In Proc. of SSS’01, pages 1–27, Atlanta, Geor-

gia. Georgia Institute of Technology.

Hillier, B. (1996). Space is the machine. Cambridge Uni-

versity Press, UK.

Hillier, B. and Hanson, J. (1984). The social logic of space.

Cambridge University Press, UK.

Janowicz, K., Scheider, S., and Adams, B. (2013). A geo-

semantics ﬂyby. In Reasoning Web. Semantic Tech-

nologies for Intelligent Data Access, pages 230–250.

Springer.

Kaplan, S. (1995). The restorative beneﬁts of nature: To-

ward an integrative framework. Journal of environ-

mental psychology, 15(3):169–182.

Neches, R., Fikes, R., Finin, T., Gruber, T., Patil, R., Sena-

tor, T., and Swartout, W. R. (1991). Enabling technol-

ogy for knowledge sharing. AI Mag., 12(3):36–56.

Nutter, J. T. (1998). Epistemology. In Shapiro, S., editor,

Encyclopedia of Artiﬁcial Intelligence. John WyleyS.

KDIR2014-InternationalConferenceonKnowledgeDiscoveryandInformationRetrieval

416

Park, T. (1993). The nature of relevance in information

retrieval: An empirical study. Library Quarterly,

63(3):318–351.

Rinaldi, A. M. (2008). A content-based approach for docu-

ment representation and retrieval. In Proc. of the 8th

ACM DocEng ’08, pages 106–109, New York, NY,

USA. ACM.

Saracevic, T. (1996). Relevance reconsidered. In Ingw-

ersen, P. and Pors, N., editors, Proc. of CoLIS’96,

pages 201–218. The Royal School of Librarianship.

Swanson, D. (1986). Subjective versus objective relevance

in bibliographic retrieval systems. Library Quarterly,

56(4):389–398.

Vakkari, P. and Hakala, N. (2000). Changes in relevance

criteria and problem stages in task performance. J. of

Documentation, 56(5):389–398.

Velarde, M. D., Fry, G., and Tveit, M. (2007). Health ef-

fects of viewing landscapes–landscape types in en-

vironmental psychology. Urban Forestry & Urban

Greening, 6(4):199–212.

SubjectivityandObjectivityinUrbanKnowledgeRepresentation

417