Aposentu: A Social Semantic Platform for Hotels

Gavina Baralla

, Simona Ibba

and Riccardo Zenoni

Department of Electrical and Electronic Engineering, University of Cagliari, Piazza D’Armi, Cagliari, Italy

Logica Solution srl, Via Giacomo Benucci, 30, Terni, Italy

Keywords:

Tourism, Complex Network, Sentiment Analysis, Social Semantic Tool, Hotel Revenue Management.

Abstract:

Tourism business has become competitive and dynamic and it is essential to adapt both customers’ satisfaction

and to market’s changing needs. A hotel owner faces three big challenges: he must look to attract new guests

to its location, manage his hotel in the best performant way and has to adapt their online marketing strategy

using several tools such as online travel agencies (OTA) or meta-search engines website or other players from

the sharing economy. All these channels can complicate the hotelier’s life, making the tourism market and

by removing in some areas the prices according to the seasons and the availability. In addition, consumer

generated content (CGCs) inﬂuence the market and the revenue management making the web reputation even

more important. This paper presents Aposentu, an innovative tool which integrates all the required components

in order to successfully manage the hotel. The platform will be cloud computing technology based and it will

show a proper dashboard with a lot of innovative functionalities. By using semantic tools, sentiment analysis,

complex network metrics, the platform will allow the hoteliers to become more competitive in the tourism

industry. Moreover, administrative complexity will be reduced and that will facilitate the management of

accommodation.

1 INTRODUCTION

Running and managing a hotel is a complex task. It

requires the ability to be ﬂexible, the capability to un-

derstand the context which they are doing business in

and to recognize the need of the travelers who look for

the best possible accommodation. Furthermore high

competition is one of the main challenges in this sec-

tor especially in those tourist destinations where, in a

small area, you can ﬁnd a lot of hotels.

Over the years a lot of software solutions have

been implemented in order to replace old-fashioned,

paper-based methods and to automatize hotel tasks

such as booking, front-end and back-end ofﬁce activ-

ities, marketing practices, accounting and so on. That

kind of tool is well known as Property Management

System, PMS and the ﬁrst implementation dates back

to 1980.

Nowadays, tourism business has become compet-

itive and the use of appropriate dynamic tools is es-

sential to keep up with the time and to make the dif-

ference compared with competitors. According to a

recent statistic (Global Property Management Soft-

ware Market Research Report 2017), the PMS market

is expected to increase by 6,99% for the period 2017

through 2022. Currently this industry is highly frag-

mented in terms of proposed solutions, often PMS are

not web-oriented (because they had been developed

before the coming of the Internet) and sometimes old

tools have been adapted to web functionality with bad

performance and by using different standards.

Tourism market is increasingly becoming web

based with the development of huge numbers of ap-

plications and websites focusing on reservation and

reputation systems. Nowadays the tourist is used to

looking for reviews before booking or buying some-

thing, Ye et al. (Ye et al., 2011) investigated the inﬂu-

ence of user-generated content to hotel online book-

ing. Gretzel et al. (Gretzel and Yoo, 2008) analyzed

the use and the impact of online travel reviews written

by consumers, they called these reviews Consumer

Generated Contents (CGCs). Furthermore, Kim et al.

(Kim et al., 2015) pointed out the importance of us-

ing social media reviews in order to manage the hotel

performance.

In addition, the online Travel Agency, OTA, provide

B2B2C services in which both the owner of the web-

site and sellers offer their products. The customer

beneﬁt sfrom this marketplace that offers multiple

kind of online booking services which help users to

Baralla G., Ibba S. and Zenoni R.

Aposentu: A Social Semantic Platform for Hotels.

DOI: 10.5220/0006511702690274

In Proceedings of the 9th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (KDIR 2017), pages 269-274

ISBN: 978-989-758-271-4

ﬁnd a hotel, buy a ticket for air travel, rent cars. At

the same time these services are oriented to different

client typologies and provide competitive prices.

In this complex scenario the hoteliers need an efﬁ-

cient software which include some different applica-

tions that often are proposed by several online ser-

vices. This multiplicity could amplify further the

work of hotelier. In order to resolve this complex-

ity we propose Aposentu, an innovative dashboard for

managing a hotel. Aposentu is the sardinian name

of the welcoming room, often dedicated to guests, in

typical sardinian house. This room represents the ﬁrst

form of reception in Sardinia.

Aposentu is a smart dashboard, completely cloud-

based on semantic tools, sentiment analysis and in

which we want to apply complex network metrics,

the platform will help hoteliers to meet tourists needs,

improve their economy facing the dynamic changing

tourist environment.

The paper is structured as follows: Section 1

provides an introduction with papers motivation,

Section 2 presents related works, in Section 3 the

proposed approach is shown. Section 4 describes the

platform architecture, ﬁnally conclusions and future

works are given in Section 5.

2 RELATED WORKS

In the last years we withnessed the development of

many applications based on semantic techniques and

projected to improve the tourist market.

An important overview of the use of knowledge

management techniques and the applications such an

approach to a tourism sector is presented by Cooper

(Cooper, 2006). He discussed how to transform in-

formation and data into capabilities for the tourism

sector.

However Shaw et. al, (Shaw et al., 2011) pre-

sented the concept of service-dominant logic and

showed how important it is to involve hotel customers

in the process of co-production and co-creation of

services. Ontologies are a useful tool to deﬁne the

relationships among concepts in a speciﬁc domain.

For instance Hontology (Chaves et al., 2012) is a

multilingual ontology for the accommodation sec-

tor that reuses some concepts extracted from Dbpe-

dia.org

and Schema.org

. Hontology could be used

in many applications that include information visuali-

sation and extraction or text annotation in an accomo-

dation context.

http://wiki.dbpedia.org/

http://schema.org/

The use of intelligent agents and artiﬁcial intel-

ligence allows the development of personalized so-

lutions. This assumption is the basis of Sem-Fit

(Garc

ıa-Crespo et al., 2011), a semantic hotel recom-

mendation system that, using fuzzy logic techniques,

takes into account consumers’ experience. However

Zhang et al. (Zhang et al., 2016) presented a senti-

mental analysis tool based on graph method, which

is applied for semantic classiﬁcation of the user re-

views collected from e-commerce websites. This ap-

proach can be joined to the typical metrics of complex

network. In order to identify competitive sets for ho-

tels useful to help hotels better position themselves

based upon eWOM (electronic word-of-mouth mes-

sages) Xiang et al. (Xiang et al., 2017) classiﬁed ho-

tel properties based on guest experiences along with

satisfaction ratings in hotel online reviews. The re-

sults of this study described the structure of tourism

industry.

Finally the system proposed by Tatiya et al.

(Tatiya and Vaidya, 2017) generates recommenda-

tions taking into account the categorical preferences

of present user and the feedback/comments of the past

users.

In our work we analyze the consumers’ experience

point of view and included in this study also all data

from main platforms of booking.

3 THE PROPOSED APPROACH

New web channels have made tourism market ex-

tremely competitive sometimes erasing season-based

prices and accommodation availability often caus-

ing price dispersion and differentiation also between

OTAs (Clemons et al., 2002). Conversely, a hotelier

decides room rates only on the basis of its experience.

The proposed platform aims at providing a tool, mod-

ular structured and cloud computing based, able to in-

tegrate different functionality and services.

The system will calculate daily a competitive price for

the offered services by considering different factors:

(i) tourist numbers, (ii) competitors average price, (iii)

web reputation, (iv) ﬁxed and variable costs, (v) hotel

rating, (vi) reservation frequency.

We want this tool to be a solution to the on-going

management hotel issues by integrating innovative

marketing perspective with web reputation aspects.

By using a smart dashboard a hotelier will have a

single interface to monitor all web channels and at

the same time he will be able to continually supervise

sales record.

The innovative system will allow the management

of internal data and their integration with external,

i.e. data competitors related to the same temporal pe-

riod and retrieved by web scraping techniques or data

coming from social networks.

The information obtained will be useful to design ad-

hoc advertising campaigns. The system will be imple-

mented following a modular architecture and it will

have:

• a dashboard which integrates in a ﬂexible way all

useful information according to the user proﬁle

and his role and capability;

• a web scraping algorithm to pinpoint data com-

ing from other booking platforms; but also a mod-

ule to manage and gather internal data. Informa-

tion will also consumer generated content derived

from web reviews;

• a Business Intelligence tool to analyze aggregated

information and based on a complex network sys-

tem;

• a Booking Engine to generate dynamic pricing;

• a tool to evaluate the web reputation and based on

sentiment analysis techniques;

• a contextual advertising module, it will be able to

record and evaluate the customer behaviour dur-

ing the web navigation (i.e. visited pages, naviga-

tion path, purchase deals etc.) in order to made an

ad-hoc targeted campaign;

• a module for social network integration.

This kind of system will allow to increase the ho-

tel turnover, manage web reputation, take over new

tourist market and become competitive.

4 PLATFORM FEATURES

We propose a modular management platform in-

tended for hoteliers and cloud based, it aims to com-

bine different services in one solution. The system

will consist of four components:

• a social semantic tool to organize tourist and ac-

commodation terminology in a semantic way;

• a contextual advertising module connected with

social networks integrated with the booking plat-

form;

• a business intelligence level will allow the com-

parison between other competitors in the same ge-

ographical area;

• a web reputation tool to evaluate users’ opinions

and hotel reviews.

All functionalities will be usable through a properly

implemented dashboard according to the user proﬁle.

Figure 1: Architecture of Aposentu.

4.1 Social Semantic Tool

An efﬁcient touristic web service needs suitable tools

to improve data management and knowledge sharing.

However a formal categorization of contents could be

insufﬁcient to interpret the user instances.

A dynamic business network could be contribute

to the evolution of an agile and collaborative book-

ing system in which the contents inserted by users

could be a sound basis to interpret better what cus-

tomers want. For instance a folksonomy allow a better

customer-centric view of a hotel’s proposition. How-

ever this popular categorization cannot be conceived

without a formal categorization that includes its capa-

bility to be trustable and a right interaction and rela-

tionship between the service components as the meta-

data of an ontology (Johnston, 2007).

Consequently Aposentu will combine tags that char-

acterize Social Web applications with an ontology to

better describe the resources speciﬁc to Semantic Web

in the context of tourism.

Our system will provide a semantic classiﬁcation

of data derived from main touristic websites, book-

ing platform including the customers comments. This

categorization will be achieved by using a double-

axis. The horizontal axis to classify consumer gen-

erated contents (CGCs) and the vertical axis in order

to solve word-sense disambiguation. Therefore we

want to integrate a formal categorization as a taxon-

omy with a folksonomy.

In order to achieve a right classiﬁcation, the data

structure of all related information will be examined

and a new innovative semantic model for tourist con-

text will be studied. The internal data or managing

information will be presented with structured or semi-

structured data to which a hierarchic classiﬁcation can

be applied. Instead in order to manage the UGCs we

will use a folksonomy and we will choose the combi-

nation of appropriate methodologies. i.e. the use of

tag cloud (Sinclair and Cardew-Hall, 2008), the ex-

traction of terms to enrich an existing classiﬁcation or

to develop a new one (Alruqimi and Aknin, 2017) or

with a social tag (Cantador et al., 2011).

We will also consider lexical database such as

WordNet

to ﬁx word-sense disambiguation.

The innovative technology will also be applied to the

connection between similar accommodation and in

the same local area in order to analyze the impact of

competitors but also the local context as well.

4.2 Targeted Advertising

A successful hotel offer has to take into account the

recent changes of Programmatic Advertising across

new media (Dawson and Lamb, 2016).

The social media tools provide an effective contextu-

ally targeted advertising solution where the opinion

of a friend or other trusted web source, indicates to

users new, appropriate and optimal products or ser-

vices. A module of user proﬁle-based marketing em-

ploys information technologies in order to recognise

the wishes of people.

An innovative tourist system has to include tools

to match the needs to the context within and around

their users (Buhalis and Foerste, 2015).

Many real time information, from mobile devices can

inﬂuence the tourist experience and help the hotelier

to understand what its prospective clients want.

This kind of system allows to revolutionise the ho-

tel offers and to create products and services dynami-

cally based on consumers’ needs.

In Aposentu we also want to include a software

module that connects the booking platform with the

most popular social networks. We will develop an al-

gorithm, based on semantic technologies to achieve a

contextually targeted advertising system.

4.3 Business Intelligence and Complex

Network

Business intelligence (Wood, 2001) models have the

capacity to study business information in order to sup-

port market management decisions. He considered

the importance of analysis of the speciﬁc context in

which a product shall be placed.

Furthermore Migu

ens et al. (Migu

ens and

Mendes, 2008) discussed how the worldwide tourist

arrivals, create a heterogeneous and directed com-

plex network in which weighted and directed net-

work measurements on its topological and weighted

https://wordnet.princeton.edu/

structure are really important to understand the tourist

ﬂows.

In Aposentu even a Business Intelligence module

will be implemented. The features of all accommo-

dation of the same level of the hotel of interest will

be analysed. In this way the impact of competitors in

a particular geographic area will be possible to ﬁnd.

A scraping tool will periodically download the prices

offered by the competitors. This information will

be processed by the business intelligence algorithm

to suggest the hotelier the best price for his offers.

We want to use metrics and measures from complex

networks to evaluate the hotel offers in the territory

through providing aggregated information about the

nature, extension and articulation of these.

The application, which will compile data collected

with special scraping algorithms, will allow the man-

agement of the information associated with each com-

petitor according to a complex network concept based

paradigm. This model will integrate ”punctual” infor-

mation, as relative to the single hotel or to the single

offer, with another type of information, of topological

nature, relating to the connections existing between

the various hotels.

These implementation, through the analysis of the

collected information, will also generate a complex

network in which the hotel offers and their reciprocal

connections will be highlighted (aggregated and with

different levels of granularity).

This representation will be visual, schematic, eas-

ily understandable, and will allow to investigate the

complex relationships of hotels.

4.4 Web Reputation and Sentiment

Analysis

User opinions and reviews on hotels on the web are a

fundamental data source in in the booking of a room.

The hotel owner must know these assessments for the

better management of resources.

We are going to include in our project a tool which

collects all opinions from main booking systems, cat-

egorizes these comments and shows information in

the best understandable way.

In our platform we want to develop a web rep-

utation tool that analyzes comments on the biggest

sites of booking (TripAdvisor, Google Reviews, Ex-

pedia, Booking, AirB&B) in order to understand the

common opinion about offers and hotel features of ac-

commodation. This component has two different pur-

poses: it monitors the use of unfair tactics on com-

ments and allows the system to show users exactly

what they wish. In this way it’s possible to keep a

high standard of service. Therefore language process-

ing algorithms will be studied.

These algorithms will interact with the text in-

serted by users in all booking platforms and, as a re-

sult of queries, will show only the offers related to ac-

commodation with appropriate characteristics or ser-

vices.

The positive or negative view of a hotel offer, the

intensity and frequency of that opinion, the emotion

with which the assessment is expressed and the rele-

vance of that deal in comparison with its geographical

area will be analysed.

5 CONCLUSIONS

This paper outlines the implementation of Aposentu,

an innovative dynamic platform, cloud based turned

at hoteliers. The system aims to integrate different

type of essential services to become competitive in

the tourism market and to keep up with time.

By using Aposentu an hotelier will successfully

manage the hotel to optimise the revenue manage-

ment. The proposed platform will also be able to

discover the users’ opinion in order to monitor and

improve web reputation.

Thanks to use of complex network metrics the sys-

tem will evaluate competitors’ offers and will provide

a useful business intelligence tool.

In order to better understand customers’ needs we

proposed a social semantic categorization that com-

bine tags with an ontology to better describe the re-

sources speciﬁc to tourism domain.

Furthermore, by using the targeted advertising

tool new guests can be attracted. Given the dynamic

context, the presented system ﬁts well with the new

internet marketing strategies. In fact all listed func-

tionalities will be implement according to a modular

architecture system and, depending on the needs, new

ones can be added as well.

REFERENCES

Alruqimi, M. and Aknin, N. (2017). Extract domain-

relevant terms from folksonomy.

Buhalis, D. and Foerste, M. (2015). Socomo marketing

for travel and tourism: Empowering co-creation of

value. Journal of destination marketing & manage-

ment, 4(3):151–161.

Cantador, I., Konstas, I., and Jose, J. M. (2011). Categoris-

ing social tags to improve folksonomy-based recom-

mendations. Web semantics: science, services and

agents on the World Wide Web, 9(1):1–15.

Chaves, M., Freitas, L., and Vieira, R. (2012). Hontology:

a multilingual ontology for the accommodation sector

in the tourism industry.

Clemons, E. K., Hann, I.-H., and Hitt, L. M. (2002).

Price dispersion and differentiation in online travel:

An empirical investigation. Management science,

48(4):534–549.

Cooper, C. (2006). Knowledge management and tourism.

Annals of tourism research, 33(1):47–64.

Dawson, P. and Lamb, M. (2016). Enhanced success with

programmatic social advertising. In Programmatic

Advertising, pages 103–110. Springer.

Garc

ıa-Crespo,

A., L

opez-Cuadrado, J. L., Colomo-

Palacios, R., Gonz

alez-Carrasco, I., and Ruiz-

Mezcua, B. (2011). Sem-ﬁt: A semantic based

expert system to provide recommendations in the

tourism domain. Expert systems with applications,

38(10):13310–13319.

Gretzel, U. and Yoo, K. H. (2008). Use and impact of on-

line travel reviews. Information and communication

technologies in tourism 2008, pages 35–46.

Johnston, K. (2007). Folksonomies, collaborative ﬁltering

and e-business: is enterprise 2.0 one step forward and

two steps back? Electronic Journal of Knowledge

Management, 5(4).

Kim, W. G., Lim, H., and Brymer, R. A. (2015). The ef-

fectiveness of managing social media on hotel perfor-

mance. International Journal of Hospitality Manage-

ment, 44:165–171.

Migu

ens, J. and Mendes, J. (2008). Travel and tourism: Into

a complex network. Physica A: Statistical Mechanics

and its Applications, 387(12):2963–2971.

Shaw, G., Bailey, A., and Williams, A. (2011). Aspects

of service-dominant logic and its implications for

tourism management: Examples from the hotel indus-

try. Tourism Management, 32(2):207–214.

Sinclair, J. and Cardew-Hall, M. (2008). The folksonomy

tag cloud: when is it useful? Journal of Information

Science, 34(1):15–29.

Tatiya, R. and Vaidya, A. (2017). Semantic-based service

recommendation method on mapreduce using user-

generated feedback. In Proceedings of the Interna-

tional Conference on Data Engineering and Commu-

nication Technology, pages 131–142. Springer.

Wood, E. (2001). Marketing information systems in tourism

and hospitality small-and medium-sized enterprises:

A study of internet use for market intelligence. Inter-

national Journal of Tourism Research, 3(4):283–299.

Xiang, Z., Schwartz, Z., and Uysal, M. (2017). Market in-

telligence: Social media analytics and hotel online re-

views. In Analytics in Smart Tourism Design, pages

281–295. Springer.

Ye, Q., Law, R., Gu, B., and Chen, W. (2011). The inﬂu-

ence of user-generated content on traveler behavior:

An empirical investigation on the effects of e-word-of-

mouth to hotel online bookings. Computers in Human

Behavior, 27(2):634–639.

Zhang, S., Song, W. W., Ding, M., and Hu, P. (2016). A

multi-semantic classiﬁcation model of reviews based

on directed weighted graph. In International Confer-

ence on Web Information Systems Engineering, pages

424–435. Springer.