Intelligent Content Management System for Tourism Smart Mobility:

Approach and Cloud-based Android Application

Alexander Smirnov

, Alexey Kashevnik

, Nikolay Shilov

Sergei Mikhailov

, Oleg Gusikhin

and Harry Martinez

SPIIRAS, 39, 14 Line, 199178, St. Petersburg, Russia

Research and Advanced Engineering, Ford Motor Company, 20300 Rotunda Drive, 48121, Dearborn, Michigan, U.S.A.

Keywords: Smart Vehicle, Connected Vehicle, Artificial Intelligence, Smart Mobility, Tourism, Recommendations.

Abstract: Intelligent content management systems have become more popular over the last few years in the tourism

industry due to the significantly increasing impact on revenue. Such systems are the part of the smart mobility

concept. Smart mobility allows tourists to become more comfortable with transportation in an unknown city

by providing interesting information about places seen during their trip. Traditional taxies provide quick

transportation from the point A to point B but people sometimes are interested in seeing attractions during

their trips and are willing to spend more time and money to do so. Integration of the smartphone application

with the vehicle infotainment system provides opportunities of new smart services construction that is based

on information from vehicle sensors and Internet connection as well as utilization of in-cabin infrastructure

and communication with the driver (suggesting the route preferred by the tourist, speed while going around

attractions, adjust the temperature, music, and etc.). The paper presents an approach to smart mobility system

development and its evaluation by showing the tourist video information about attractions around.

1 INTRODUCTION

Information technologies fill different aspects of our

lives with innovative services supporting various

activities. Development of artificial intelligence

technologies, in turn, improve the level of support

(e.g., Barolli and Enokido, 2017).

One of such supporting solutions is intelligent

personal assistant. Such assistants are basically

software applications that are capable of helping

people in their everyday activities (Santos et al.,

2018). Intelligent personal assistants can access

information from databases and other sources to

guide people through different tasks, applying

learning mechanisms to acquire new information

about user preferences.

Tourism is a rapidly developing area of human

activity, where information technologies are used

very intensively. Information on attractions, transport

schedules, electronic maps, feedback from other

tourists, etc., all this information can be acquired from

various sources and used for improving the tourist

experience (Li et al., 2017; Ukpabi and Karjaluoto,

2017).

However, the amount of available information is

becoming overwhelming and its manual processing is

getting difficult. This is where application of artificial

intelligence with its capability of personalized

information analysis can be beneficial. Such systems

can not only extract required information from

various sources, but also adapt to tourists’ behavior,

interests and preferences (Xiao et al., 2017; Wadekar

at al., 2017; Li and Cao, 2018; Smirnov et al., 2018).

Integration of such intelligent tourist support

systems with vehicles would produce a new

experience. The in-vehicle services could provide for

more precise positioning, speed and direction

estimation, route prediction based on the navigation

system, etc. The paper presents the intelligent tourist

support system utilizing in-vehicle services.

Described earlier, the approach (Smirnov et al., 2018)

has been implemented in an Android-based system.

The paper concentrates on intelligent information

extraction from various sources and cloud-based

application architecture.

The structure of the paper is as follows. The

approach is explained in section 2. The information

extraction and application structure are presented in

section 3. Section 4 explains implementation details

426

Smirnov, A., Kashevnik, A., Shilov, N., Mikhailov, S., Gusikhin, O. and Martinez, H.

Intelligent Content Management System for Tourism Smar t Mobility: Approach and Cloud-based Android Application.

DOI: 10.5220/0007715304260433

In Proceedings of the 5th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2019), pages 426-433

ISBN: 978-989-758-374-2

and demonstrates the user interface. Finally, remarks

are given in the conclusion.

2 STATE-OF-THE-ART

Today information availability through Internet

makes it possible for tourists to search and discover a

significant amount of interesting information on their

own. This gives a rise to the individual tourism, when

tourists travel by themselves and use various services

as assistance and guiding.

The tourist sights (palaces, museums, parks, etc.)

are also searching for new ways of using information

technologies to attract more tourists, providing for

smart audio tours, interactive information kiosks, and

other (Smirnov et al., 2017a; Giuseppe et al., 2014).

As a result it has become more interesting to spend

time at the attractions than just casting a glance,

taking a picture and walking away.

Usage for smart services and mobile applications

makes it possible for the tourists to better plan their

routes, see more attractions and spend more time

there compared to “old fashioned” means such as

booklets and paper maps (Li and Liew, 2015;

Modsching et al., 2007). Such new services help

tourists to conduct some research before travelling to

better plan the trip and find convenient travelling

means (such as public transport, taxi services, etc.)

Transport providers, in turn, also recognise the

appearance of new opportunities and increasing

amount of individual tourists. Most of the tourist

cities have a network of hop-on-hop-off buses, which

drive by a predefined route from one attraction to

another, and tourists can enter and leave such bus

without a need to buy a ticket every time. Some cities

have 2 or 3 day tickets for public transport that are

usually enough for covering the travelling needs of

individual tourists.

This makes it possible to increase the utilization

by each customer, spread the use over the whole day,

and allow the operator to get more utilization out of

each bus, instead of optimizing for peak periods.

Taxi companies do not only introduce mobile

apps that make it possible to order taxi for a

predefined price without a need to negotiate with a

driver who may not speak the tourist’s language, but

also seek to new tourist-oriented business models.

For example, UberTour launched by Uber

combines the idea of the Uber taxi (when one can

reserve a car using a mobile app) and hop-on-hop-off

buses (with a possibility to enter the car and leave it

at some pre-defined stops associated with attractions).

The conducted analysis of the available tourist-

oriented services and mobile applications

(Ambrosino et al., 2010; Price, 2015) indicates that

there is a tendency to provide the tourists with

proactive information support taking into account not

only the current location but also personal

preferences, weather and other context information

(Cowen, 2015; Smirnov et al., 2014b; Gerhardt, 2015;

Staab et al., 2002; Hasuike et al., 2015). It is still an

actual task to develop a system that would be capable

to :

 collect information about attractions from

different sources and recommend the tourist the

best for him/her attraction images and

descriptions;

 generate recommended attractions and their

visiting schedule based on the tourist and

region contexts and attraction estimations of

other tourists. The tourist context characterizes

the situation of the tourist, it includes his/her

location, co-travelers, and preferences; the

region context characterizes the current

situation of tourist location area, it includes

such information as weather, traffic jams,

closed attraction, etc.

 propose different transportation means for

reaching the attraction;

 update the attraction visiting schedule based on

the development of the current situation.

3 AN APPROACH

The conceptual scheme of the smart mobility system

is presented in Figure 1. Based on the schedule

analysis (e.g., hotel check-out before 11AM and

flight at 08PM) the system proposes a tour that would

fit the schedule and finish at the airport. Alternatively,

the person can schedule a tour of his/her own. The

tour takes into account the person’s preferences

(preset and revealed via collaborative filtering

techniques) and the current situation at the location

(season, weather, traffic jams, etc.) based on the

earlier developed personalized tourist assistance

service TAIS (Smirnov et al., 2017; Mikhailov,

Kashevnik, 2018). The person can modify the

recommended tour and share with other tourists. The

driver picks up the tourist(s) at the predefined time

and follows the route loaded into the car’s navigation

system automatically via the car connectivity means.

Personal electronic devices (such as smartphone)

could be used during the tour for narration, imagery

and video synchronized with the vehicle’s location

Intelligent Content Management System for Tourism Smart Mobility: Approach and Cloud-based Android Application

427

Figure 1: Conceptual scheme of the smart mobility system.

and speed. The support is based on the ad-hoc “on-

the-fly” combination of the available information

pieces (pre-recorded narration, images, etc.) in the

personalized context-aware manner.

4 CONTENT MANAGEMENT

The proposed intelligent content management system

consists of smartphone application and cloud

application. Smartphone application is aimed at on-

the-fly intelligent information provision to the tourist

while the cloud application is utilized for attraction

information accumulation and smart services

construction such as context determination and

recommendation generation.

4.1 Smartphone Application

The reference model of the developed smartphone

application is presented in Figure 2. The tourist

interacts with the graphical user interface that

provides the following functions.

 Attraction page generation function is aimed at

presenting the user the accessible on/offline

attraction database content about the chosen

attraction. Attraction page includes the text

information, audio and/or video description,

and or pictures related to the attraction.

 Map generation function is used by the tourist

to see his/her location and accessible attraction

around in the interactive map. The

OpenStreetMap service is used for these

purposes.

 Path calculation function provides the optimal

path (by time or by distance) from the tourist

location to the chosen tourist attraction.

 Route calculation function is used to suggest

the tourist acceptable attraction visiting route

based on the current situation in the region

incorporating his/her preferences.

 Proactive recommendation function allows the

mobile application to generate personalized

tourist context-aware recommendations and

propose in proactive mode attractions that are

better to attend for him/her.

 When the tourist reaches their attraction the

intelligent narration function is used to present

him/her information about the attraction. Based

on the vehicle speed and importance of the

attraction the length of the narration should be

calculated (length of the text, length of the

audio, or length of the video).

Such functions as attraction page generation,

proactive recommendations, and intelligent narration

are the core functionality of the content management

system and implemented in the tourist smartphone

and they do not require internet connection. The

extended functions such as map generation, path

calculation, and route calculation are implemented in

the cloud application.

The database module is used to store information

about attractions in the offline attraction database in

the tourist smartphone and provide this information to

the graphical user interface. Offline attraction

database is synchronized with the cloud application

where it is generated based on accessible information

from OpenStreetMap, Wikipedia, and Expert

Knowledge. The database module supports the

following functions.

 Attraction description storing that incudes

information in text form.

 Attraction media that includes information in

audio, and/or video, and/or pictures.

 List of attractions function that generates the

list of attractions in the tourist location area.

VEHITS 2019 - 5th International Conference on Vehicle Technology and Intelligent Transport Systems

428

Figure 2: Reference Model of the Smartphone Application.

 Text information about attractions as soon as

their location is stored in SQLite database. The

media information that includes audio, video,

and pictures is stored as a file in the filesystem.

4.2 Cloud Application

The reference model of the developed cloud

application is presented in Figure 3. The model

consists of several services that are implemented in

Docker containers: Database Manager Service, Nginx

Web Service, Reverse Geocoding Service,

Recommendation Service, Route Planning Service.

The Database Manager Service automatically

creates an attraction database based on the

information from OpenStreetMap and Wikipedia.

The OpenStreetMap is used to select the list of

attractions in the location region while the Wikipedia

contains the text description and pictures of the

attractions. The Database Manager Service supports

the following functions.

 Attraction viewing uses the json format that is

used by the smartphone application to get

attraction information.

 Attraction management function supports the

adding, removing, and updating of attractions

in the attraction database as well as their text &

media description.

 Region creation function allows the creation of

new regions and the construction of the

attraction database for this region based on

information from OpenStreetMap and

Wikipedia.

 Offline DB creation function provides the

ability to create the SQLite database that can

be downloaded by the tourist for local

utilization.

The Attraction Database is implemented using the

PostgreSQL database management engine.

The Nginx Web Service provides the ability to

manage the attraction database by tourism experts.

The service provides the web interface for the

attraction management function of the Database

Manager Service. It supports the adding, removing,

and updating of attractions in the attraction database

as well as their text & media description.

The Reverse Geocoding Service is implemented

to determine the tourist location city by coordinates.

It uses the GeoNames web service for this purposes.

The Recommendation Service implements the

artificial intelligence functions of the developed

content management system. The main function of

the service is to organize the attraction visiting list

based on the user preferences and context situation in

the region. The collaborative filtering approach is

used for recommendation generation utilizing the

tourist experience with content management system

in the past.

The Context Service provides context situational

data to the content management system. At the

moment of service, it takes into account the following

information: weather, traffic jams, user context. All

this information is then incorporated into the

intelligent recommendations generation for the

tourist.

The Route Planning Service functions are: route

creation and path creation. Path creation is used for

shortest path creation between the tourist location and

the chosen attraction. The route creation is the

intelligent function that finds the attractions that are

the most interesting for the tourist based on his/her

preferences and context situation in the region

(weather). Based on this list and context situation

(traffic jams) the service generates the route that is

Tourist

Graphical User Interface

1. Attraction page

2. Map generation

3. Path calculation

4. Route calculation

5. Proactive recommendations

6. Intelligent narration

Offline Attraction

Database (SQLite

database & media files)

Database Module

1. Attraction description

2. Attraction media

3. List of attractions

OpenStreetMaps

Map

Cloud

Application

Intelligent Content Management System for Tourism Smart Mobility: Approach and Cloud-based Android Application

429

Figure 3: Reference Model of the Cloud Application

most suitable for the tourist to move from one

attraction to another.

5 IMPLEMENTATION

The developed intelligent content management

system shows the tourist information about

attractions in a proactive mode (Figure 4 and Figure

5). It supports the function of database download and

utilizes it to provide the tourist information without

the Internet connection. On the left side of the screen,

the list of ordered attractions is shown. The tourist can

see the pictures related to his/her location and the

region map. The middle of the screen shows the

offline attraction database manager function that

shows what the information the database system has

at the moment for St. Petersburg and Bologna. The

right screen shows the information about the

attraction chosen by the tourist.

When the tourist is riding in a vehicle the system

automatically determines the movement direction and

recognize the side where the attraction is located and

notifies the tourist regarding the direction to the

attraction (left, right) and presents images and

audio/video/text information about it (Shilov et al.,

2018). Based on information in the database, the

tourist preferences, and context situation (speed) the

system plays pre-recorded narration.

To test the performance of the proposed system

the attraction visiting plan has been generated, based

on data from the of St. Petersburg city in the Android

emulator. The computer specification is the

following: Intel(R) Core(TM) i7-8700 CPU

@3.20GHz process, 16 Gb RAM, 1 TB HDD.

Docker Container

OpenStreetMap

Wikipedia

Attraction Database

(PostgreSQL)

Experts

Database Manager Service

Docker Container

Nginx Web Service

1. Vue application (css & js)

2. Attraction Media

1. Attraction viewing (json)

2. Attraction management

3. Region creation

(OSM, Wikipedia)

4. Offline DB creation (SQLite)

Docker Container

Context Service

1. Weather

2. Traffic jams

3. User context

Docker Container

Reverse Geocoding Service

1. City determination by

coordinates

2. Traffic jams

City Database

(PostGIS)

Docker Container

Route Planning Service

1. Route creation

2. Path creation (GraphHopper)

Docker Container

Recommendation Service

1. Ordered attraction visiting list

Application

VEHITS 2019 - 5th International Conference on Vehicle Technology and Intelligent Transport Systems

430

Figure 4: Screenshots of smartphone application: the main screen and offline database download window.

Figure 5: Smartphone application showing direction to the attraction (left) and attraction description (right).

Each attractions contains general information,

five keywords, and three images. As a result, sets of

1 000, 5 000, 10 000, 15 000 and 20 000 attractions

have been formed. Figure 6 contains information

about overall database size for each attraction set and

Figure 7 represents offline database download time.

The database includes textual information in the

SQLite format and graphical data such as images in

the separate files. At the moment attraction database

for St. Petersburg contains only 1 000 attractions. 20

000 attractions is highly enough for storing

information about European city capital. This kind of

database requires ~ 4 Gb of memory on a smartphone.

A small drawback of such amount of data is quite a

long time downloading offline databases (around of

35-40 minutes for 20 000 attractions dataset). This

issue can be addressed by loading the offline database

in a background thread and processing available data

in parts. As Figure 8 shows the attraction textual data

retrieving applicable for this amount of information.

Intelligent Content Management System for Tourism Smart Mobility: Approach and Cloud-based Android Application

431

Figure 6: Relationship between offline attraction database

size and attraction count.

Figure 7: Relationship between offline attraction database

downloading time and attraction count.

Figure 8: Attraction search time dependency on attraction

count in the considered region.

6 CONCLUSIONS

The paper presents an intelligent content management

system for tourism smart mobility. The presented

system is developed for the tourist smartphone and

cloud service. The core functions are supported by the

smartphone directly and the tourist does not need to

have internet connectivity. Some functions are

supported by the cloud service and the tourist will

need to have an internet connection to use them.

For the future work authors are going to develop

the tour guiding application using Applink API,

running on the tourist’s Android-based device,

implementing the proposed approach, and capable of

communicating with the proposed cloud service.

Development of the intelligent application prototype

supporting interactive context-dependent guidance is

continued with additional features to be implemented

(higher level of personalization, tour planning,

feedback accumulation for collaborative filtering).

ACKNOWLEDGEMENTS

The paper is due to the project sponsored by the Ford

University Research Program.

REFERENCES

Ambrosino, G, Nelson, J D, Bastogi, B, Viti, A,

Romazzotti, D Ercoli, E. 2010. The role and

perspectives of the large-scale Flexible Transport

Agency in the management of public transport in urban

areas. In Ambrosino, G, Boero, M, Nelson, JD and

Romanazzo, M (Eds), Infomobility Systems and

Sustainable Transport Services, Rome: ENEA, pp. 156-

165..

Barolli, L., Enokido, T. (Eds.). 2017. Innovative Mobile

and Internet Services in Ubiquitous Computing. In

Proceedings of the 11th International Conference on

Innovative Mobile and Internet Services in Ubiquitous

Computing (IMIS-2017). Advances in Intelligent

Systems and Computing, Vol. 612. Springer. 978p.

Cowen, B. 2015. A personal tour guide – almost

everywhere – for $9.99 or less! URL:

http://www.johnnyjet.com/2015/01/a-personal-tour-

guide-almost-everywhere-for-9-99-or-less/.

Gerhardt T. 2015. 3 Ways Multi-Modal Travel is Tricky for

App Developers. URL: http://mobilitylab.org/

2015/03/10/3-ways-multi-modal-travel-is-tricky-for-

app-developers/.

Giuseppe, M., Michela, B., Matteo, F., Chiara, F.,

Alessandro, M., Stefano, M., Fabiana, R. 2014..

QRCODE and RFID integrated technologies for the

enhancement of museum collections. In: Euro-

Mediterranean Conference, Springer, Cham, pp. 759-

766.

Hasuike, T., Katagiri, H. Tsubaki, H. Tsuda, H. 2015.

Interactive approaches for sightseeing route planning

under uncertain traffic and ambiguous tourist’s

satisfaction. In: H. Eto (ed.) New Business

Opportunities in the Growing E-Tourism Industry,

Hershey, PA: Business Science Reference, pp. 75-96.

Li, R. Y. C., Liew, A. W. C. 2015. An interactive user

interface prototype design for enhancing on-site

museum and art gallery experience through digital

technology. Museum Management and Curatorship,

30(3), pp. 208-229.

Li, Y., Cao, H. 2018. Prediction for Tourism Flow based on

LSTM Neural Network. Procedia Computer Science,

129, 277-283.

Li, Y., Hu, C., Huang, C., Duan, L. 2017. The concept of

smart tourism in the context of tourism information

services. Tourism Management, 58, 293-300.

Mikhailov S., Kashevnik A. 2018. Smartphone-based

tourist trip planning system: a context-based approach

to offline attraction recommendation. In 13th

International Scientific-Technical Conference on

Electromechanics and Robotics “Zavalishin’s

Readings”. Vol. 161.

2000

4000

0 5000 10000 15000 20000 25000

Database size, Mb

Attraction count

0 5000 10000 15000 20000 25000

Download time, sec

Attraction count

1000

2000

0 5000 10000 15000 20000 25000

Fetching time, ms

Attraction count

VEHITS 2019 - 5th International Conference on Vehicle Technology and Intelligent Transport Systems

432

Modsching, M., Kramer, R., Ten Hagen, K., Gretzel, U.

2007. Effectiveness of mobile recommender systems

for tourist destinations: A user evaluation. In Intelligent

Data Acquisition and Advanced Computing Systems:

Technology and Applications, 2007. IDAACS 2007.

4th IEEE Workshop on, IEEE, pp. 663-668.

Price E. 2015. Travel apps that can replace your tour guide.

URL: http://www.cntraveler.com/stories/ 2015-02-

24/travel-apps-that-can-replace-your-tour-guide.

Santos, J., Rodrigues, J. J., Casal, J., Saleem, K., Denisov,

V. 2018. Intelligent personal assistants based on

internet of things approaches. IEEE Systems Journal,

Vol. 12, No. 2, pp. 1793-1802.

Shilov, N., Kashevnik, A., Mikhailov, S. 2018. Context-

Aware Generation of Personalized Audio Tours:

Approach and Evaluation. In: Speech and Computer,

20th International Conference on Speech and Computer

(SPECOM 2018), Leipzig, Germany, Springer Vol.

11096. P. 615–624.

Smirnov A., Shilov N., Gusikhin O. 2017a. Cyber-

Physical-Human System for Connected Car-Based e-

Tourism. In: Proceedings of 2017 IEEE Conference on

Cognitive and Computational Aspects of Situation

Management, Savannah, GA, USA, 27-31 March 2017,

IEEEComSoc.

Smirnov A., Shilov N., Gusikhin O. 2018a. Context-

Dependent Guided Tours: Approach and Technological

Framework. In: 3rd International Scientific Conference

“Intelligent Information Technologies for Industry”,

Advances in Intelligent Systems and Computing, Vol.

874, Springer, pp. 43-50.

Smirnov, A., Kashevnik, A., Ponomarev A. 2017b.

Context-based infomobility system for cultural heritage

recommendation: Tourist Assistant—TAIS. In

Personal and Ubiquitous Computing, Springer,

Heidelberg. P. 297–311.

Smirnov, A., Ponomarev, A., Shilov, N., Kashevnik, A.,

Teslya, N. 2018b. Ontology-Based Human-Computer

Cloud for Decision Support: Architecture and

Applications in Tourism. International Journal of

Embedded and Real-Time Communication Systems

(IJERTCS), 9(1), 1-19.

Smirnov, Kashevnik, A. Ponomarev, A., Shilov, N. Teslya,

N. 2014. Proactive Recommendation System for m-

Tourism Application. In: B. Johansson, B. Andersson,

N. Holmberg (eds.) Perspectives in Business

Informatics Research, Proceedings of the 13th

International Conference on Business Informatics

Research (BIR 2014), Springer, LNBIP 194, pp. 113-

127.

Staab, A., Werthner, H., Ricci, F., Zipf, A., Gretzel U.,

Fesenmaier D.R., et al. 2002. Intelligent systems for

tourism. In: IEEE Intelligent Systems, No 6, pp. 53-64.

Ukpabi, D. C., Karjaluoto, H. 2017. Consumers’ acceptance

of information and communications technology in

tourism. In: A review. Telematics and Informatics,

34(5), 618-644.

Wadekar, P., Agrawal, S., Paunikar, A., Dobarkar, S.,

Chaudhary, S., Wankhede, S., Students, B. E. 2017.

Implementation of Mapper Reducer for Intelligent

Tourism System. International Journal of Engineering

Science, 5337.

Xiao, Z., Sen, L., Yunfei, F., Bin, L., Boyuan, Z., Bang, L.

2017. Tourism Route Decision Support Based on

Neural Net Buffer Analysis. Procedia Computer

Science, 107, 243-247.

Intelligent Content Management System for Tourism Smart Mobility: Approach and Cloud-based Android Application

433