LOCATION INTELLIGENCE SERVICES FOR MOBILES USING

RUBY ON RAILS AND JQUERYMOBILE

Alfio Costanzo, Alberto Faro and Concetto Spampinato

Department of Electrical, Electronics and Computer Engineering, University of Catania, Catania, Italy

Keywords: Location Intelligence, Fuzzy Systems, Semantic Web, Location based Services, Mobile Computing.

Abstract: The current applications for mobiles provide information about Location Based Services (LBS) of user

interest that don't take into account the current status of the services neither they take into account the real

time conditions of the traffic network or of the weather. Aim of the paper is to show how location

intelligence services may be implemented by an open source solution that is able to help the activities and to

support the decisions of the mobile users taking advantage from all the information collected by the sensors

located on the field and from the business data stored on the disparate data stores that may be of interest of

the urban/metropolitan mobility, security and logistics. Fuzzy logic and semantic web technologies are

taken into account to improve the current LBS applications. A case study developed using Ruby on Rails

and JQueryMobile illustrates how such a web service may work in practice.

1 INTRODUCTION

The current applications for mobiles provide

information about Location Based Services (LBS) of

user interest that don't take into account the current

status of the services neither they take into account

the current conditions of the traffic network or of the

weather. Overcoming such limits is not easy since

this needs the availability of information outside the

control of the service providers, such as the travel

time to reach the destination depending on the

current traffic flows, the current availability of

parking vacancies and so on.

Also, the mobile applications don't support m-

commerce, i.e., commercial transactions carried out

on-line by mobile users. Implementing real time and

m-commerce services is only the first step to

activate advanced LBSs. In fact, it is also important

to provide the users with the information that best

fits their needs. For example, a cheaper park may be

suggested with lower priority if there is another park

that is more close to the destination in case of

raining. Another step is the one of integrating the

administrative Data Bases (DBs) resident on

separate computers relevant for LBS so that the user

may be informed on all the services potentially

available at the urban/metropolitan scale.

For this reason, the future mobile application

should be powered more and more by artificial

intelligence techniques, thus giving rise to location

intelligence services of practical user interest (Faro,

2011a). Although many intelligent mobile services

may be theoretically offered to the users, as the ones

proposed in the field of the intelligent transportation

systems (ITS) (McCubbin, 2003), they are rarely

offered in practice to the users. Often, the services

offered are not conceived to support the user

decisions or to optimize the user activities but to

provide the users with average information that give

a moderate help to their activities.

Moreover, ITS services are mainly distributed by

proprietary solutions that don't take advantage from

the information resident on the disparate DBs of

potential user interest. For example, sudden traffic

congestions or dangerous situations are not reported

to the interested users, service reservation cannot be

carried out on-line, neither the users are advised

when products of their interest are available on some

store.

Aim of the paper is to show how location

intelligence services may be implemented by an

open source solution that is able to help decisions

and activities of the mobile users taking advantage

from all the information collected in the data stores

that may be of interest of the urban/metropolitan

mobility.

In the paper we assume that the area is provided

with sensors that monitor in real time the traffic

763

Costanzo A., Faro A. and Spampinato C..

LOCATION INTELLIGENCE SERVICES FOR MOBILES USING RUBY ON RAILS AND JQUERYMOBILE.

DOI: 10.5220/0003940707630771

In Proceedings of the 8th International Conference on Web Information Systems and Technologies (WEBIST-2012), pages 763-771

ISBN: 978-989-8565-08-2

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

conditions and the weather. Also, we assume that

business information about availability, vacancies

and cost of urban/metropolitan services are stored in

some server connected to internet.

Thus, the paper does not deal with the basic

monitoring technologies widely studied in the

literature, e.g., (Faro, 2008, 2011b)], but it is

specifically devoted to illustrate how to integrate all

the relevant information available on the net to

support the main use cases of the mobile users

depending on their current position. Fuzzy logic

rules and metadata technologies are used to help the

user mobility and her/his mobile business activities.

The Ruby on Rails (RoR) (Hartl, 2011)

framework is adopted to develop the web service

since it allows the designer to organize the

application as a collection of use cases that can be

reused for similar tasks (Faro, 1998, 2033a).

Moreover, RoR is provided with a powerful

language, i.e., Ruby, that facilitates the

implementation of: a) the fuzzy rules that address

user mobility and aid their decisions, and b) the

procedures to access the metadata layer that

integrate the disparate DBs. Other two languages

may be also used in RoR to facilitate the

implementation of LBS applications: a) Java scripts

to exchange with mobiles information geo-

referenced on Google Maps, and b) JQueryMobile

(Bai, 2011) to convey such information in a user

friendly format that may be visualized, without any

modification, on PCs, tablets and mobiles.

Section 2 shows how simple fuzzy logic and

Horn Logic rules may be implemented by Javascript

and JQueryMobile to improve the location based

services. Section 3 discusses how the use of

JQueryMobile and Javascript allow us to use the

metadata technology to favour the integration of the

proprietary DBs of interest of mobile users. The

advantages of implementing the Location

Intelligence services by using RoR and JQuery

Mobile will be discussed in section 4 by a small case

study dealing with a prototypical web application,

called WiCity, currently under test at our University,

that illustrates how an user can connect her/his

mobile to an RoR server to be informed by a suitable

interface, developed by JQueryMobile, on some

basic LBSs concerning mobility (parks, gas stations

and traffic congestions), health services (pharmacies

and first aid services), events/places of tourist

interest, and on the routes to reach by car the chosen

destination from the current user position by taking

into account the current traffic flows and weather

conditions. When discussing this prototype we will

outline how the methodological issues pointed out in

the previous sections may be used to implement

effective location intelligence services for citizens

and tourists.

2 IMPROVING LBS BY FUZZY

AND HORN CLAUSES

Location based services may be considered as a sort

of generalization of the Intelligent Transportation

Systems. The latter are mainly dedicated to improve

mobility and logistics activities of the mobile users,

the former aim at supporting such activities taking

into account also security, commerce and business

requisites. Thus, in LBS environment it is important

to reach the destination in the minimum time, but

also to avoid accidents or congested areas.

Analogously, the user may decide to follow some

non minimal path to reach the destination if this is

done in more safe conditions depending on the

weather conditions. As well as, the user may be

interested in paying the parks depending on the real

parking time rather than paying in advance basing

on some forecast of this time.

Although the rules that support such LBS

requirements are very simple, the current LBS

applications are mainly conceived as information

systems that provide the users with general

information while they are walking or driving. Thus,

in the following we will show how the LBS

effectiveness may be improved by using suitable

fuzzy logic rules (Wang, 2001) and Horn clauses

expressed in Prolog (Wielemaker, 2009).

2.1 Nearest Services and Safe Walking

To help the users to choose the most suitable nearest

services we should return to them a Google Map on

her/his mobile that shows the current user position

and two circles, one, let say C

, with a radius of few

hundreds of meters and the second, let say C

, with a

radius of about one kilometre, containing the

markers of the services located in such areas. The

former circle should point out the services that are

reachable by walking, whereas the second should

point out the ones reachable by car.

In principle, one can define the circle radius

following a very simple rule, i.e., the services

located in the smaller circle are recommended to the

walking people if the distance dist from the current

user position is such that dist < 300 meters. This

constraint would become dist < 1000 meters if the

user is driving a car.

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

764

However, if there are no services available

within such circles, the user might accept a

moderately greater distance to find services of

interest. Of course, an arbitrary modification of such

distance is not acceptable, whereas it is fair to

inform the user on how much a certain increase of

the circle radius may causes a corresponding

decrease of her/his satisfaction of the solution

provided.

The computation of the radius corresponding to

the user expectations is straightforward in the fuzzy

logic framework in case we have to consider only

one condition that may influence the notion of

nearest. Indeed, for example, assuming that the rule

is "if the user is not young, then the user would like

to have the required service very close", and that the

user is 28 years old, from the fuzzy sets in fig.1c we

have that the evidence that she/he is not young is

given by the membership not [μ

young

(28 years old)],

i.e., 0.8, and consequently the most suitable radius is

the x-coordinate of the barycentre of the area M1 in

fig.1a, i.e., 175 meters.

The computation of the best radius d is a few

more complicated if we wish to take into account

more conditions together, plus the current user

position. In fact, the RoR application should be

provided with the fuzzy set of each condition and

with an algorithm to combine the radii derived from

the various conditions. To show how this can be

done, let us assume that our rule is: "if the user is not

young and it is cloudy, then the user would like to

have the required service very close". In this case, if

the people is 28 years and the sky is partially cloudy

(e.g., cloudiness degree = 0.4), we may derive the

maximum distance as the x-coordinate of the

barycentre of the two masses M1 and M2 in fig.1a.

The former represents the distance to be suggested

considering only the age, the latter refers only to the

cloudiness. Approximately it is 190 mt. Of course, if

the and contained in the rule is substituted by or, the

radius decreases since it is the x-coordinate of the

barycentre of the mass M1 or M2 that has the greater

membership, i.e., about 175 mt.

Thus, to solve the above problem, the RoR

application should know the current time and

weekday, as well as the user age and health status,

and the traffic and weather conditions measured by

dedicated sensing infrastructures.

How to measure the travel times for each street

has been widely analyzed by the authors in other

papers, whereas, in the case study, we will show

how the current weather conditions may be obtained

by connecting the RoR web application to Yahoo.

For what concerns the walking time to

destination we may assume that it depends on the

distance between the current user position and the

destination as suggested by Google Maps. But, in

case there are dangerous areas to avoid, the RoR

web application should inform the users about

alternative paths to reach either the destination or

safer locations. However, indicating alternative

paths using Google Maps is not a trivial job since

the routes suggested by Google Maps are based on

average conditions that cannot be modified easily by

the programmer. A solution of this problem is the

one of displaying the alternative pedestrian routes

like the minimum time driving routes using Google

Maps but in the more elaborated way discussed in

the next section.

NEAREST

500

300

1

CLOUDY

YOUNG

30

40

60

MIDDLE

AGED

ELDERL



0.6

0.2

cloudiness

degree

0.6

0.4

Figure 1: Fuzzy sets associated with the words: nearest,

young, middle aged, elderly, and cloudiness, where μ [0,1]

is the membership degree to such words of the values of

the definition domain, e.g., μ

young

(28 years old) = 0.2.

2.2 Safe and Fast Driving in the Traffic

The computation of the minimal route connecting

two intersections is certainly instrumental to find the

best route between two addresses. It may be

age (years)

distance d from the

user position (meters)

LOCATIONINTELLIGENCESERVICESFORMOBILESUSINGRUBYONRAILSANDJQUERYMOBILE

765

obtained by any program that is able to find the best

path connecting two points of a graph (Kumari,

2010). In (Faro, 2011a) we have suggested a

Mobility Based Prolog (MBP) Program, since it may

be used with little modifications to find also other

routes of interest of the user as the ones to find the

service that are nearest to the destination.

Also, this MBP program shows very satisfactory

time performance and allows us to update the travel

time of a street between two adjacent intersections x,

y from to t

to t

by simply retracting the fact

travel_time (x, y, t

) and asserting the new fact

travel_time(x, y, t

) in the Prolog knowledge base. In

this way, any accident or congestion or work in

progress involving a street may be immediately

taken into account by the program.

Another advantage of using Prolog is that a

simple generalization of the MBP program allows us

to solve the most relevant logistics problem, i.e., the

one to compute the minimal Hamiltonian cycle of

the graph associated to the traffic net, i.e. the cycle

that starting from an address will come back to the

initial address in a minimum time by traversing a set

of prefixed intermediate points.

Moreover, the previous fuzzy rules may be easily

written in Prolog, thus making possible that the

MBP program uses the membership values to

express if the routes or the mentioned Hamiltonian

cycles found are very (enough/few) fast

(fluent/slow). As a consequence, if the user is

interested only to fast routes to destination, the

program may suggest the route only if μ

fast

route

) >

0.8, otherwise the user is invited to postpone her/his

travel.

2.3 Displaying the Best Routes

A suitable graphical interface that is compatible with

the one used frequently by the user is very important

for the usability of the web service (Giordano,

2002). For this reason we aim at informing the user

on the best path to destination or on the best cycle to

distribute or collect goods, by representing the

traffic network as a graph superimposed to the

Google Maps in such a way that its arcs coincide

with the streets and the nodes with the street

intersections.

Thus, after having identified the triples

travel_time (x, y, t) for every adjacent intersection

pair and having found the best route by using he

mentioned MBP program we should draw on Google

Maps the best path from a source intersection S to

the destination intersection D by the subsequent

drawing of linear traits connecting the adjacent

nodes traversed during the path. However, the lines

of such drawing don't correspond necessarily to

those of Google Maps since the links between nodes

are not always linear segments, neither we have a

database containing the adjacent nodes with their

geographical coordinates. Thus, we have two

problems: a) to find the geo-coordinates of all the

intersections, and b) to draw the links between

adjacent nodes like the ones of Goole Maps.

To solve the first problem it would be enough to

use the function "x AT y" that gives the coordinates

of the intersection between the road x and y. But,

this function is available for US, and not for all the

countries. Thus, our first problem is to find an

alternative way to compute all the intersection geo-

coordinates in the urban/metropolitan area.

Fortunately, this can be accomplished as follows: a)

pass to the API Directions of Google Maps the

names of the pair of streets (x, y) that have an

intersection to find the route that allows us to reach

by walking the initial address of x from the last

address of y (or in some case the initial address of x

to the initial address of y), and b) extract the

geographical coordinates of the first marker that

contains in its info window the name of the route y,

thus finding the geo-coordinates of the intersection.

However, as pointed out above, the knowledge

of the intersections together with their

geo/coordinates is not enough to find the optimal

route from any source address ad

to any destination

address ad

. Indeed, to solve the problem we have to

execute the following further tasks:

• to find the adjacent intersections. This can be

obtained by using the API Directions by verifying

for each pair of intersections if they are connected

by one step link, and

• to compute for any address ad

the set AD

out

(ad

)

of intersections that can be reached in one step from

and the set AD

(ad

) of intersections that allow

us to reach in one step ad

. These sets can be

computed by using Directions to find the

intersections around ad

that can be reached

driving in one step from ad

and the intersections

around ad

that can be reached

by walking in one

step from ad

. Of course, we have to exclude in the

latter case the routes not allowed to drivers.

Finally, we have to compute, e.g., by using the

mentioned MBP program, all the routes connecting

any intersection in AD

out

to any intersection in AD

The best route is the one that is obtained by

minimizing, for any intersection belonging to AD

out

(ad

) and to AD

(ad

), the travel time T consisting

of the following three terms:

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

766

T = t(ad

, AD

out

(ad

) ) + t(AD

out

(ad

) , AD

(ad

) )) + t(AD

(ad

), ad

)

(1)

To draw the same intersection links that will be

drawn by Google Maps, we use again Directions to

draw the connections between the adjacent

intersections of the best route by simply requiring

that such route is obtained by using repeatedly

Directions to draw the best route between any pair

of adjacent nodes belonging to the best route.

3 DATA INTEGRATION USING

OWL-LIKE METADATA

The use of proprietary DBs is still convenient today

to manage the data warehouse of any organization.

But, simple commercial transactions and pure

information tasks push more and more for the use of

standard formats, such as RDF based DBs (Powers,

2003), especially in the LBS framework. Indeed, the

data stores based on OWL, that is an extension of

RDF, or even on XML favour the integration of data

belonging to different organizations. For example,

the 'public' part of the data of an organization could

be mapped in OWL and sent to a central server

where such data will be available for all the users

through a standard interface.

Alternatively, the XML/OWL data could remain

on the servers of the organizations if the central

server is able to carry out distributed queries to

collect the data useful for the mobile user. This will

favour the updating of the data and the system

reliability.

The data, in standard format, could be also stored

on the mobiles, even if this solution is suitable only

for data that are few dependent on time, otherwise

their frequent updating may interfere with the

normal operations of the mobile. The technologies

available on the market allows us to implement all

the above solutions not only to support the

centralized or the distributed access to the DBs, but

also to facilitate the mapping of the relational DBs to

triple stores, or the production of novel OWL DBs

from scratch, e.g., (Allemang, 2011), (Bonomi,

2007), (Faro, 2003b) and (Zhai, 2008). For example,

software environments such as Protègè

(Knubblauch, 2004) are suitable to design novel

RDF stores, whereas servers such as Sesame

(Broekstra, 2002) may be used to implement

centralized RDF stores available to web users. The

use of JQueryMobile facilitates the access to the

DBs from mobiles either directly or with the help of

the central information server (David, 2011).

Developing the information server as an RoR

application allows us to design the web service as a

collection of use cases. This will improve the

verification, the test and the maintenance of the

software especially when the work flow of the

application is complicated for the presence of

several cooperating actors.

4 CASE STUDY

This case study illustrates how we connect a mobile

to an RoR web application, called WiCity, that

includes all the technical issues discussed in the

previous sections. The interested reader may

download the software from code.google.com/

p/query-mobile-unict/source/browse/.

Fig.2 shows the WiCity architecture, where the

user mobiles are connected to a central RoR server.

Currently, all the needed information is stored in the

MySQL tables of the RoR server. We are also

developing a distributed version of WiCity where

the RoR server will use both local databases and the

remote XML/OWL data stores resident on the

proper remote servers (i.e., Sesame server). This

architecture will favour data integration, data

privacy and data updating according to the methods

outlined in sect.3.

Figure 2: Current centralized WiCity architecture. It is

conceived to support future integrations of XML/OWL

data stored on different remote servers.

Fig.2 points out that, currently, the user mobiles

may use only data stored on the RoR server,

whereas, in the future remote data may be used too.

In particular, data integration will be obtained by

connecting local and remote XML/OWL data stores

through messages exchanges between the web RoR

application stored on the RoR server and the Sesame

servers at distant service points.

The mobile interface, developed by

JQueryMobile, currently consists of some icons

User

Mobile

RoR

Server

MySQL

Tables

XML

OWL

Sesame

Server

Sesame

Server 1

XML

OWL

XML

OWL

. . . . .

LOCATIONINTELLIGENCESERVICESFORMOBILESUSINGRUBYONRAILSANDJQUERYMOBILE

767

concerning information on mobility and health

services, and on the best routes to reach the

destination from the current user position (fig.3.left).

Also, events/places of general interest are provided

to tourists and citizens (fig.3.right).

Figure 3: The interface icons of WiCity are chosen from a

predefined list: main interface (on the left) and events of

general interest (on the right).

The events of general interest consist of icons

chosen by the user from the list of available LBSs.

In particular, the one on the left_top in fig.3.rigth is

connected to an RoR process implemented on the

server that is able to access the weather metadata of

Yahoo. After having processed the JSON file

received from Yahoo such process extracts the

weather conditions of the area in which the mobile is

located, i.e., cloudy and 17 °C in fig.4.left.

Such data are not only useful to inform the user

on the current weather conditions but also to modify

the fuzzy sets associated to the concept nearest

service. The basic mobility and heath services (e.g.,

parks, gas stations, pharmacies and first aid points)

may be accessed either in alphabetical order to get

relevant information such as address, location on the

map, opening hours and so on (see fig.4 right), or by

the mentioned fuzzy facilities.

Figure 4: Weather info and localization of a pharmacy of

user interest chosen from a predefined list.

Currently, only fuzzy facility that take into

account the current traffic and weather conditions

are available; they are denoted as services nearest to

my current position. For example, fig.5 shows how

WiCity points out the pharmacies that may be

reached by walking or by car displaying the

corresponding markers within a circle of a suitable

radius obtained by using the mentioned fuzzy rules.

Figure 5: Pharmacies that may be reached by car (on the

left) and by walking (on the right) taking into account the

weather info and the user position.

To use the facilities of WiCity, any user should

be registered. In the information related to her/his

profile, WiCity includes automatically the items

inserted by the user so that they are available for the

other users of her/his community as shown in fig.6.

In this ways we should obtain two benefits:

• to encourage the users to insert information

useful for their community, and

• to avoid that they insert deliberately wrong

information.

The use of the registered information to provide the

users with e-government and e-commerce services

(e.g., certificates, event tickets, etc.) is for future

works.

Figure 6: Registration form (on the left) and list of the

items inserted by the user (on the rigth).

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

768

In fig.7 we show the WiCity mobile interface

obtained by using Flash Builder by Adobe (Gassner,

2010). The effort for developing this version of

WiCity is certainly lower than the one needed to

develop an RoR application powered by

JQueryMobile, but the organization of the data is not

so effective as the one supported by RoR that is

based on the well known paradigm Models-Views-

Controllers (Hartl, 20011).

Figure 7: Some WiCity snapshots in Flash Builder.

One positive feature of the Flash Builder

applications is that it may use XML-like files, as the

ones outlined in fig.8, stored locally or on a remote

server.

Figure 8: The XML file used by the Flash Builder

application installed on the user mobile.

A more powerful facility of using metadata may

be obtained by using the RoR framework. As an

example, WiCity makes possible the geo-

localization of the services by geo-markers whose

info windows are filled with the data extracted from

XML/OWL metadata as shown in fig.9 and fig.10.

Figure 9: Geo-localization of points that are reachable by

walking and by car. These points are related to the event

described in the window on the bottom. The data of the

event are taken from the metadata.

Finally, let us clarify in fig.11 how the

Directions API of Google Maps is useful to find the

intersections between two routes x and y in order to

obtain the function (x AT y) currently not available

for many countries.

Indeed, in fig.11.left the intersection between the

road named 'costarelli' and the one named 'del

toscano' is not pointed out by using Directions to

connect 'costarelli' and 'del toscano' with the option

by driving, whereas it is pointed out when it is

executed with the option by walking. Indeed, often

there is no intersection between two routes by using

Directions with the option by driving, since as

shown in fig.11.left, Directions takes into account

the one way streets. On the contrary, the information

contained in the info window of the function

Directions with the option by walking (e.g.,

fig.11right) allows us to discover the marker and

related geo-coordinates associated to the intersection

between the roads.

LOCATIONINTELLIGENCESERVICESFORMOBILESUSINGRUBYONRAILSANDJQUERYMOBILE

769

Figure 10: The info window associated to the points of

interest is filled with the data extracted from the metadata.

How to obtain all the intersections and to derive

the matrix that gives the adjacent intersections of

each traffic network intersection is outside the scope

of the paper and will be discussed in detail in future

works. Once such matrix is obtained, the

methodology proposed in sect.2.3 will allow us to

display on the user mobiles the current best route to

destination.

Figure 11: Route connecting the initial and final addresses

of two streets in search of the marker associated to their

intersection by driving (on the left) and by walking (on the

right).

5 CONCLUSIONS

The paper has widely discussed how the current

LBS applications may be improved by using fuzzy

rules and semantic technologies. The solutions

proposed would have the expected high impact since

the information will be based on real time updated

data stores.

Another expected positive feature of the

proposed web services is that the chosen

implementation approach favours the integration of

the disparate database available at metropolitan scale

thus opening concrete opportunities to develop

mobile-commerce and mobile-business applications

of wider utility.

Further studies are planned to evaluate how much

location intelligence should be embedded in the LBS

applications to really improve the location services

offered to the users.

Issues like decision support systems based on

fast data mining techniques (Faro, 2011c) and fast

image pre-processing, e.g., (Cannavò, 2006)

(Crisafi, 2008), for supporting people recognition

and people flow control in case of emergency should

take advantage from the possibilities of having

JQueryMobile based PDAs that are able to suggest

timely convenient alternatives in both security and

logistics fields to mobile users.

The applications presented in the paper are

currently under development within a project called

K-Metropolis supported by our Region to favour the

transition towards the knowledge society with the

aim of improving the level of competitiveness of the

local economic system.

Further applications of the proposed technologies

are also planned at our University to control physical

processes that may influence the people security and

the environmental quality such as control systems

that alert the drivers on the overflowing of a river by

indicating suitable escape routes, or emergency

systems that inform timely the policeman in case a

high pollution is affecting a certain area of the sea

(Spampinato, 2010).

REFERENCES

Allemang D., Hendler, J., 2011. Semantic Web for the

Working Ontologist: Effective Modeling in RDFS and

OWL, Elsevier Ltd, Oxford, 2011

Bai G., 2011. JQueryMobile. Packt Publishing, 2011

Bonomi A., Rondelli A., Vizzari G., Stride S.., 2007. An

Ontology Driven Web Site and its Application in the

Archaeological Context, 2nd International Workshop

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

770

on Ontology, Conceptualization and Epistemology for

Software and System Engineering, 2007

Broekstra J. et al., 2002. Sesame: a generic architecture for

storing and querying RDF and RDF Schema. LNCS

Vol.2342, 2002.

David M., 2011. Developing Websites with

JQueryMobile. Focal Press, 2011

Cannavò F., Nunnari G., Giordano D., Spampinato C.,

2006. Variational Method for Image Denoising by

Distributed Genetic Algorithms on GRID

Environment. Proc. Int. Workshops on Enabling

Technologies: Infrastructure for Collaborative

Enterprises, WETICE '06. IEEE, 2006

Crisafi A., Giordano D., Spampinato C., 2008. GRIPLAB

1.0: Grid Image Processing Laboratory for Distributed

Machine Vision Applications. Proc. Int. Workshops on

Enabling Technologies: Infrastructure for

Collaborative Enterprises, WETICE '08, IEEE, 2008

Faro, A., Giordano, D., 1998. StoryNet: an Evolving

Network of Cases to Learn Information Systems

Design. IEE Proceedings SOFTWARE, 145(4), 119-

127, 1998

Faro, A., Giordano, D., 2003. Design memories as

evolutionary systems: socio-technical architecture and

genetics, Proc. IEEE International Conference on

Systems Man and Cybernetics, Vol.5, 4334-4339,

IEEE, 2003

Faro, A, Giordano, D. and Musarra, A., 2003. Ontology

based intelligent mobility systems. Proc. IEEE

Conference on Systems, Man and Cybernetics,

Washington, Vol.5, 4288-4293, IEEE , 2003.

Faro, A., Giordano, D., Spampinato, C., 2008. Evaluation

of the traffic parameters in a metropolitan area by

fusing visual perceptions and CNN processing of

webcam images. IEEE Transactions on Neural

Networks, vol.19(6), 1108-1129, IEEE, 2008

Faro, A., Giordano, D., Spampinato, C., 2011. Integrating

location tracking, traffic monitoring and semantics in a

layered ITS architecture, Intelligent Transport

Systems, IET, Vol.5(3), 197-206, IET, 2011

Faro, A., Giordano, D. and Spampinato, C., 2011.

Adaptive background modeling integrated with

luminosity sensors and occlusion processing for

reliable vehicle detection. IEEE Transactions on

Intelligent Transportation Systems. Vol. 12(4), 1398-

1412, IEEE, 2011.

Faro, A., Giordano, D., Maiorana, F., 2011. Mining

massive datasets by an unsupervised parallel clustering

on a grid: Novel algorithms and case study. Future

Generation Computer Systems, Vol. 27(6), 711–724,

2011.

Giordano, D., 2002. Evolution of interactive graphical

representations into a design language: a distributed

cognition account, International Journal of Human-

Computer Studies, Vol. 57(4), 317-345, 2002

McCubbin, R. P. Staples, B. L. and Mercer, M. R. 2003.

Intelligent Transportation Systems Benefits and Costs:

2003 Update, Mitretek Systems, Inc., Washington.

Gassner D., 2010. Flash Builder 4. Wiley, 2010

Hartl M., 2011. Ruby on Rails 3, Addison Wesley, 2011.

Knubblauch H. et al., 2004. Protege OWL Plugin (an open

development environment for semantic web

applications. Proc. ISWC 2004, LNCS Vol.3298, 229-

243, 2004

Kumari, S. M. and Geethanjali, N., 2010. A Survey on

Shortest Path Routing Algorithms for Public Transport

Travel, Global Journal of Computer Science and

Technology, Vol. 9(5), 2010.

Powers,S. 2003 Practical RDF, O’Reilly Media

Spampinato, C., Giordano, D., Di Salvo, R., Chen-Burger,

Y. H. J., Fisher, R. B., Nadarajan, G., 2010. Automatic

fish classification for underwater species behavior

understanding, in Proceedings of the first ACM

international workshop on Analysis and retrieval of

tracked events and motion in imagery streams,

ARTEMIS ’10, ACM, 45-50, 2010.

Wang P. P., 2001. Computing with words. Wiley

Inderscience, 2001

Wielemaker, J. Hildebrand, M. and Ossenbruggen, J.

2009. Using Prolog as the fundament for applications

on the semantic web, hcs.science.uva.nl/projects/SWI-

Prolog/ articles/mn9c.pdf, 2009.

Zhai, J., Jiang, J., Yu, Y. and Li, J., 2008. Ontology-based

Integrated Information Platform for Digital City, IEEE

Proc. Of Wireless Communications, Networking and

Mobile Computing, WiCOM '08, IEEE, 2008.

LOCATIONINTELLIGENCESERVICESFORMOBILESUSINGRUBYONRAILSANDJQUERYMOBILE

771