=Paper=
{{Paper
|id=Vol-1630/paper2
|storemode=property
|title=A Recommendation Engine Based On Social Metrics
|pdfUrl=https://ceur-ws.org/Vol-1630/paper2.pdf
|volume=Vol-1630
|authors=Ofelia Cervantes,Francisco Gutierrez,Ernesto Gutierrez,J. Alfredo Sanchez,Muhammad Rizwan,Wan Wanggen
|dblpUrl=https://dblp.org/rec/conf/semweb/CervantesGGSRW15
}}
==A Recommendation Engine Based On Social Metrics==
<pdf width="1500px">https://ceur-ws.org/Vol-1630/paper2.pdf</pdf>
<pre>
              A Recommendation Engine based
                    on Social Metrics

    Ofelia Cervantes1 , Francisco Gutiérrez1 Ernesto Gutiérrez1 , J. Alfredo
            Sánchez1 , Muhammad Rizwan2 , and Wan Wanggen2
               1
                Universidad de las Américas Puebla, Puebla, México
          2
           Institute of Smart City, Shanghai University, Shanghai, China
         ofelia.cervantes@udlap.mx, francisco.gutierrez@udlap.mx,
         ernesto.gutierrezca@udlap.mx, alfredo.sanchez@udlap.mx,
                  rizwan@shu.edu.cn, wanwg@staff.shu.edu.cn



      Abstract. Recommender systems have changed the way people find
      products, points of interest, services or even new friends. The technology
      behind recommender systems has evolved to provide user preferences and
      social influences. In this paper, we present a first approach to develop a
      recommendation engine based on social metrics applied to graphs that
      represent object’s characteristics, user profiles and influences obtained
      from social connections. It exploits graph centrality measures to elab-
      orate personalized recommendations from the semantic knowledge rep-
      resented in the graph. The graph model and selected graph algorithms
      for calculating graph centralities that are the core of the recommender
      system are presented. Semantic concepts such as semantic predominance
      and similarity measures are adapted to the graph model. Implementation
      challenges faced to solve performance issues are also discussed.

      Keywords: recommender systems, social metrics, graph models, mas-
      sive graph processing


1   Introduction

The widespread availability of products and services through web-based and
mobile applications makes it difficult for end users to select the right item, ac-
cording to their preferences. A recommender system must make use of different
sources of information for providing users with suggestions of items that better
correspond to their expectations. Those sources can include user preferences,
item descriptions or social information.
    Several approaches have been proposed to tackle the problem of selecting
automatically the list of items that really contributes to satisfy the needs of end
users. Approaches based on demographics or modeling user profiles are oriented
to exploit user features and preferences for filtering available choices. The main
challenge of this approach is to create the user profile from scratch. Some systems
invite users to select their preferences from a predefined list of categories or
allow the system to extract their profiles from other applications. Other systems
2

register every user action to dynamically build models based upon the user’s
behavior. Some research works focus on collecting ratings made by users that
have evaluated the product or service providing a first-hand perception of the
quality of the evaluated item. Other approaches have focused on describing the
main features of every item and trying to match them with user requirements.
    Modeling social networks using graphs has opened opportunities for exploring
new alternatives for implementing recommender systems. Social metrics, such as
flow centralities calculated on graph-based models provide interesting measures
to represent the semantic predominance of concepts featuring users’ preferences
as well as item characteristics. A first proof of concept was accomplished and
reported in [6]. A more detailed description of the model is presented here as
well as implementation problems that were solved to provide a complete recom-
mendation engine that can be integrated in recommendation applications.
    The remainder of the paper is structured as follows: Section 2 presents related
work. Then, Section 3 introduces the proposed graph-based recommendation
model. Section 4 discusses selected graph algorithms used to calculate social
metrics, particularly flow centralities. Next, Section 5 describes the framework
developed in the prototype and the performance challenges we had to overcome.
Finally, in Section 6 we report the results we have obtained thus far and discuss
future work.


2     Related Work

Traditionally, recommender systems are classified into the following categories:
demographic, content-based, collaborative filtering, social-based, context-aware
and hybrid approaches. However, the ever increasing amount of information
available in social media enables the development of knowledge-based recom-
mender systems [4]. The rich knowledge that has been accumulated in social
media can be exploited to improve recommendation outcomes, enhance user
experience and to develop new algorithms. In addition to the classification of
recommender systems, Zanker [24] illustrates the fundamental building blocks
of a recommender system, out of which we highlight three: user model, commu-
nity (social network), and a recommendation algorithm. Thereby, we present a
brief overview of related work regarding knowledge-base recommender systems
from building blocks perspectives.


2.1   User Modeling

In order to personalize recommendations, it is necessary to know information
about each user. User models are representations of users’ needs, goals, prefer-
ences, interests, and behaviors along with users’ demographic characteristics [19].
Several user modeling approaches have been proposed, from typical weighted
vectors to domain ontologies. In [1] the authors defined a user model based on
fuzzy logic and proposed an approach to infer the degree of genre presence in
                                                                                    3

a movie by exploiting the tags assigned by the users. In [24] the authors pre-
sented a simple attribute-value pair dictionary to model the user through the
explicit elicitation of user requirements. A richer user model is presented in [10],
where the authors used a machine learning process to capture the user profile
and context into a domain ontology.
    Our work tries to balance between simple [24] and complex models [10] with
the goal of having an efficient but still rich user model. Other works, like Canta-
dor et al. and Moahedian et al. [5, 14], are similar to our proposed user model,
since we use tags and keywords to build a lax ontology.


2.2   Recommendation Algorithms and Techniques

There is a wide range of recommendation algorithms and techniques. They vary
according to data availability, recommender filtering type as well as user and
object representation. Several methods have demonstrated to have an acceptable
performance such as: association rule learning [24], Bayesian networks [8], nearest
neighbors [2], genetic algorithms [13], neural networks [3], clustering [20], latent
semantic features, among others that can be found with more detail in [4].
    In this work, we use graph metrics commonly used in social network anal-
ysis [21]. We propose semantic social network analysis that integrates semantic
methods of knowledge engineering and natural language processing with classic
social network analysis. Advantages of semantic social network analysis are: its
knowledge foundation and its non probabilistic nature. In contrast, one disad-
vantage is its computational cost. Therefore, enhancement techniques are needed
to process graph metrics more efficiently.


2.3   Information Extracted from Social Networks

Recommender systems are creating unique opportunities to assist people to find
relevant information when browsing the web, and making meaningful choices
with the success of emerging Web 2.0, and various social network Websites. In
[7], the author has proposed a novel approach for recommendation systems that
uses data collected from social networks.
     Wang et al. [22] studied the problem of recommending new venues to users
who participate in location-based social networks (LBSNs) and propose algo-
rithms that create recommendations based on: past user behavior (visited places),
the location of each venue, the social relationships among the users, and the sim-
ilarity between users.
     Ye et al. [23] exploited the social and geographical characteristics of users and
locations/places to research issues in realizing location recommendation services
for large-scale location-based social networks. They observed the strong social
and geospatial ties among users and their favorite locations/places in the system
via the analysis of datasets collected from Foursquare.
     Similar to our work, Savage [18] investigated the design of a more complete,
ubiquitous location-based recommendation algorithm that is based on a text
4

classification problem. The system learns user preferences by mining a person’s
social network profile. The author also defined a decision-making model, which
considers the learned preferences, physical constraints, and how the individual
is currently feeling.
    We can state that novel approaches rely mainly on the fusion of information
inferred from a user’s social network profile and other data sources (e.g. mobile
phone’s sensors). In this sense, it is necessary to develop new strategies that
produce recommendations from rich but still incomplete information.


3     Graph Model

Our recommendation engine is based on a graph representation of users and
objects of interest linked through concepts (denoted as terms). Figure 1 shows
the graph model where every node falls in one of three categories: User, Term
or Object of Interest, and every edge represents the semantic relation between
nodes: Predominance, Similarity or Friendship. Every Term node of the graph
in Figure 1 acts as a semantic descriptor of both: Users and Objects of Interest.
In other words, every user and every object are correspondingly described by the
terms linked to them. In general, users are described by their tastes, preferences,
and interest (user model) whereas objects are described by tags and keywords
(object model). In this manner, when a term is shared between a user and an
object, it shows the possibility that the user could be interested in that particular
object, even though the object had never been seen or rated by user.
    A graph-based representation allows us to apply graph algorithms (e.g. so-
cial metrics) to discover topological features, key relationships, and important
(prestigious) nodes. Then, with these features we can make relevant recommen-
dations to users, such as suggesting friends or places. Therefore, the foundation
of our recommender system relies on a knowledge base constructed from both:
a user model (see section 3.3) and an object model (see section 3.4).
    In order to construct the user and object models, we applied a linguistic
analysis over user and object text descriptions. Basically, we conducted pre-
processing (removal of stopwords and selection of most descriptive words) and
statistical linguistic analysis (using weighting schemes: tf-idf and okapi BM25)
to define a bond between text descriptions and semantic relations represented in
the graph (see section 3.2). It is possible to obtain user and object descriptions
from social networks (Facebook, Foursquare, Twitter), web pages (Wikipedia,
web search results, etc.), human experts contributions or other textual resources.


3.1   Weighted Graph Definition

Formally, we define a weighted graph G = (V, E, fE ) where V = {v1 , . . . , vn }
is a set of vertices vertex (nodes), E = {e1 , . . . , en } ⊂ {{x, y} | x, y ∈ V } is a
set of edges, and fE : E → R the function on weights for every edge. In our
recommender system V = U ∪ T ∪ O where U is the set of users, T is the set
of terms, and O is the set of objects of interest. E = P ∪ S ∪ F where P is
                                                                                 5




Fig. 1. Graph model. Node layers: User, Term, and Objects of Interest linked trough
Friendship, Predominance and Similarity edges.


the set of predominance edges, S the set of similarity edges, and F is the set
of friendship edges. Function fE is adapted according to each type of edge. For
instance, we can obtain the sub-graph of users as GU = (U, F, fF ) (see User
layer in Figure 1), the sub-graph of objects as GO = (O, S, fS ) (see Objects
of Interest layer in Figure 1), and the sub-graph of user and object profiles as
GU ∪T ∪O = (U ∪ T ∪ O, P, fP ).

3.2   Semantic Relations
In order to build the semantic relations of the graph, it is necessary to obtain
text descriptions of users and objects. As a result, we have two collections: the
users text collection (UTC) and the objects text collection (OTC), where each
text description is considered a document D in a vector space model.
    We define three types of semantic relations (edges of the graph): predomi-
nance, similarity, and friendship. Each semantic relation links different types of
nodes and has a different weighting function. Predominance is the edge between
a user or an object and a term, similarity is the edge between two objects and
friendship is the edge between two users.
    Predominance is the semantic relation between a term and a user or an object.
A term acts as a descriptor of users and objects. We define a weighting function
over the edge of predominance based on linguistic analysis. We apply Okapi
BM25 ranking function [17] to each independent document collection (UTC and
OTC) using Equation 1.
    In Equation 1, pred is the predominance of the term T in document D, Idoc
is the number of indexed documents (size of collection), Tdoc is the number of
documents containing term T , T F is the term frequency relative to document D,
DL is the document length, avgDL is the average document length among the
entire collection, K and B are free parameters (usually K = 1.2 and B = 0.75).
6



                                                                                   !
                             Idoc + 0.5                    T F ∗ (K + 1)
    pred(D,T ) = log10                        ∗                             DL
                                                                                         (1)
                             Idoc + 0.5           T F + K ∗ ((1 − B) + B ∗ avgDL )

   Similarity is the semantic relation between two objects. This measure in-
dicates the degree of affinity between objects. We apply the cosine similarity
measure (Equation 2) to obtain this value. The Similarity is calculated after the
predominance, since it relies on shared terms. Then, every object is a vector of
predominances as shown below Equation 2.
                                                             A·B
                         Similarity(A,B) = cos Θ =                                       (2)
                                                            kAk kBk

                                                                         
                ObjectA = pred(A,T1 ) , pred(A,T2 ) , · · · , pred(A,Tn )
                                                                        
                ObjectB = pred(B,T1 ) , pred(B,T2 ) · · · , pred(B,Tn )

    In Equation 2, the similarity between object A and object B is determined by
the weights of the terms they have in common. In this manner, a high similarity
value indicates a higher semantic correspondence between objects.
    Friendship is the semantic relation between two users. This measure indicates
the degree of affinity between two users. Our current model does not distinguish
between close friends, friends or acquaintances. Therefore, the users’ sub-graph
is only a friend-of-a-friend (FOAF) node-link type.

3.3    User Model
As part of the graph-based representation, users are defined as sub-graphs. A
user model is composed of two sub-graphs: user profile Gu = (u, P, fP ) and user
FOAF network GU = (U, F, fF ). Figure 2A shows the user profile network and
2B the user FOAF network. In the user profile sub-graph, each user is linked
to a set of terms that indicate tastes, preferences and interests. Tastes are gen-
eral inclinations of user towards some entities and they are generally expressed
with actions such as likes (e.g. Foursquare, check-ins and Facebook likes football,
beer, steak, coffee, etc ). Preferences are user inclinations towards taste features.
Preferences are more fine grained than tastes and are usually expressed in users’
reviews and ratings (e.g. starred reviews: I like the double espresso, I don’t like
diet soda). Interests are defined as contextual user inclinations or intentions (e.g.
I want to try Chinese food, I’m going to watch minions movie).
                               
                                pred(U,T )         Case A
                         fP = 1                      Case B                       (3)
                                  #stars − 2 ∗ 13    Case C
                               

   Our scheme to weight edges within a user profile is indicated in Equation
3. Case A occurs when only text descriptions are used; this means that terms
                                                                                 7




                Fig. 2. A)User profile and B)User FOAF Network


are weighted according to the Okapi measure (pred, as shown in Equation 1).
Case B occurs when explicit likes are found in Foursquare or Facebook. Case C
occurs when terms extracted from starred reviews are used to describe a user.
In addition to Equation 3, we use a threshold value to limit the number of
terms connected with a given user. In fact, we use the first quartile as threshold
value. An example of user profile is shown in Figure 2A, where, it is possible to
notice that a user likes football, rock and coffee, and is likely to be student. In
user friendship networks, as mentioned earlier, there are no differences among
friendship types. Then, in FOAF network all weights are equal to 1 (fF = 1). A
user FOAF network is shown in Figure 2B.


3.4   Object Model

Objects of interest are sets of items that can be of potential interest to a user.
Depending upon the application, objects of interest can be of different grain
size. For instance, they can be coarse-grained as a point of interest (POI) or
fine-grained as items inside places. An object is described by the category to
which it belongs (e.g. dog is an animal). Also, it is described by tags designed
by users or keywords found in object’s description. The object model is consists
of two sub-graphs: object profile Go = (o, P, fP ) and object similarity sub-graph
GO = (O, S, fS ). Object profile was built with data gathered from Foursquare,
Wikipedia and results from web searches. The weights of edges that link objects
and terms were calculated using the predominance formula shown in Equation
1. This means that the weight function on edges is fP = pred(O,T ) .


3.5   User Global and Local Network

In order to apply social metrics (centrality measures) and relate them to perti-
nent recommendations, we defined two networks from user perspective: a user
global network (U GNu ) and a user local network (U LNu ). User global network
is the whole graph (all nodes: users U , terms T and objects O and all edges:
similarities S, predominances P and friendships F ) centered in current user.
8

Therefore, U GNu = (U ∪ T ∪ O, S ∪ P ∪ F ) (see Figure 1). Whereas user local
network is the sub-graph defined by current user node u, term nodes adjacent to
user Tu and object nodes adjacent to terms node OT linked trough predominance
edges from user Pu and from objects PO . Hence, U LNu = (u ∪ Tu ∪ OT , Pu ∪ PO )
(see Figure 3). It is important to highlight the difference between user global
and local networks, since it will lead to different semantics interpretations when
calculating centrality measures over them.


4     Centrality Measures and Recommender Engine
Centrality measures have been used extensively in the past to exploit networks
and discover the most relevant nodes in a graph. In social network analysis
(SNA) graph centralities are used to identify the most relevant persons, com-
munities and even detect strange behaviors in the network. However, given the
takeoff of social networks, people have increased their interaction not only to
meet people and friends but to search about things they like, give their impres-
sions, and reach points of interest and objects of interest. These spatio-temporal
interactions can also be represented in a graph, thus, an accurate user profil-
ing representation can give us a great deal of insight about the user behavior.
Because of these approaches in exploiting graph measures we have explored the
use of centralities to exploit the topological structure of our user global network
and flow centrality to get the most of our weighted global and local networks in
terms of a recommender engine.

4.1   Centrality Algorithms
Centrality in graphs is widely used to measure the importance of a node in a
graph, especially in SNA [9]. Our recommender engine implements these central-




                            Fig. 3. User Local Network
                                                                                 9

ities to measure the relevance of people in the social network. Some centrality
measures like closeness and betweenness are based on the calculation of the
shortest distance to reach all other nodes in the graph. Our algorithms to calcu-
late centralities are applied to the network of persons so we can infer the most
popular nodes (degree), the capacity of a node to reach any other in the network
(closeness), and to identify the leaders interconnected within a neighborhood
in the graph (betweenness) [15]. Degree centrality is a measure that counts the
direct relationships a node has, and thus, the nodes that are in direct contact.
Closeness is defined as the inverse sum of the shortest paths to each other node
and betweenness is defined as the number of shortest paths from all vertices
to all others that pass through that node. Centrality measures are calculated
over the network at a topological level given a scale-free graph of persons. Thus,
these measures are not exploiting our weighted graph, they are applied only at a
social-network level. Terms and objects of interest can be seen as sub-graphs of
the global network that can be exploited by using flow-based centrality measures.

4.2   Flow-Centrality Algorithms
We are using flow centralities [15] to measure the betweenness, closeness, and
eccentricity in between the objects of interest, terms, and people profiles. Flow
centralities allow us to exploit the semantic relationships between the user and
the profiles of the objects of interest. Flow centralities reveal the most relevant
nodes in the graph given their weights, for instance, given a set of terms asso-
ciated to a user profile we can better understand a user preferences and give a
better recommendation.

Flow Betweenness In SNA betweenness is one of the most common referenced
centralities. Flow betweenness (see Equation4) is defined in [11] as the max flow
that passes through node xi , by the total flow between all pairs of points (pi )
where xi is neither a source nor a sink. A node with a high flow betweenness
centrality has a large influence in the network because of the flow that passes
through it. Due to the relevance of a node with high betweenness, in our rec-
ommender model, a node with high betweenness should be recommended as the
things the user cannot miss (see Figure 4).
                                      Pn
                                        j<k mjk (xi )
                             Fb (i) = Pn                                       (4)
                                          j<k mjk


Flow Closeness Closeness is just a measure of distance and is defined as the
inverse of the average distance to other vertices. A node with high flow closeness
centrality has a fast communication within all the nodes in the graph. In equation
5, flow closeness is defined as the inverse sum of the max flow to every other
resource. In our recommender model, elements with high flow closeness should
be recommended to the user as things that could be interesting, because those
weighted elements are close to the user profile (see Figure 5).
10




        Fig. 4. “You can’t miss” as a result of computing flow betweenness.



                                             1
                             Fc (i) = Pn                                      (5)
                                         j<k mjk (xi )


Eccentricity On the other hand, eccentricity is the maximum distance taking
in consideration the weighted paths of the network. Eccentricity lets us find the
nodes that are far away from the most central node in the network. In equation
6 eccentricity is defined as the maximum distance between pairs of nodes given
their maximum flow in the network. In our recommender model, an item with




         Fig. 5. “Could be interesting” as a result of computing closeness.
                                                                                     11




      Fig. 6. “Nothing else to do? Try this” as a result of computing betweenness.

                     Table 1. Centrality-based recommendations.

                        Measure          Recommendation
                        Similar nodes    What else?
                        Flow betweenness You can’t miss
                        Flow closeness   Also discover
                        Eccentricity     Be different



high eccentricity should be recommended if the user has nothing left to do (see
Figure 6).
                                                X
                             Fe (i) = maxj v       mjk (xi )                        (6)


4.3    Graph recommendations

As we have shown, our recommender model relies on the continuous computation
of predominance and similarity between items in the graph. As the graph evolves
from user interactions between the user and objects of interest, the recommenda-
tions get more accurate over time. However, in order to give recommendations,
computation of centralities is required. As shown in Table 1, we can recommend
similar items if the user is asking “What else?”, then we can show him similar
items to the recommended item. Flow betweenness is used to recommend things
the user “can’t miss” because of their relevance in the network. Flow closeness is
used to recommend central items that could be things that the user “would like
to discover”, and eccentricity is used to show items to the user that are far away
from the more central nodes in the network and could cause a “being different”
impression.
12

5        Implementation
We decided to implement the discussed model and graph measures in a weighted
graph. We explored different graph databases and graph processing frameworks,
to select the tools to build a graph processing framework that could give us the
flexibility to calculate those metrics with ease.

5.1      Census framework
We defined an architecture (see Figure 7) based on the graph model discussed
before. We named our graph processing framework “Census” which is built with
Play framework1 and is intended to have multiple instances of Signal/Collect
while processing our graph in Google Compute Engine2 . Census uses Neo4j3
Graph database to store the graph, Neo4j give us the flexibility to do queries
over the computed network through custom plugins that serve queries through a
REST API. Census process requests from Census Control which uses an orches-
trator to administrate compute requests and instances of Census in the graph.

5.2      Proof of concept
With Census we explored a first approach to implement the presented recom-
mender model. The graph database was populated with nodes of persons and




             Fig. 7. Architecture of Census, graph processing framework.
     1
       https://www.playframework.com/
     2
       https://cloud.google.com/compute/
     3
       http://neo4j.com/
                                                                                 13

                      Table 2. Centrality computation results

                          Flow       Flow       Flow
 Point of Interest                                        Predominance
                       Betweenness Closeness Eccentricity
 Centro Expositor          0.00       0.00       0.39         0.39
 Africam                   1.05       0.04       0.28         0.47
 Museo Revolución        6.00       2.08        0.46         1.22
 Convento Sta. Monica      2.21       0.73       0.46         0.67
 Museo Amparo              0.00       0.00       0.65         0.65
 Museo Regional Puebla     2.77      1.04        0.65         0.66
 Fuerte Loreto             0.00       0.00       0.78         0.78
 Casa Alfeñique           0.00       0.00       0.78         0.40
 Catedral Puebla          7.13        1.01       0.31         0.64
 Exconvento Calpan         0.00       0.00       0.40         0.40
 Convento San Gabriel      0.00       0.00      0.84          0.84
 Capilla Rosario           0.00       0.00      0.84          0.40
 Fuerte Guadalupe          0.00       0.00       0.23         0.23
 Plazuela Sapos            2.04       0.16       0.35         0.64
 Barrio Analco             0.00       0.00       0.41         0.41
 Paseo San Francisco       0.00       0.00       0.25         0.25
 Barrio Artista            0.00       0.00       0.25         0.21
 Estadio Cuauhtémoc       1.05       0.12      0.85          1.38



points of interest from the city of Puebla, Mexico, then we selected documents
from the Web to create a profile of the points of interest using our semantic ap-
proach described before. We created different users with their respective profiles
setting them with random characteristics as weights in the relationships of the
graph. We calculated similarities between those points of interest and the terms
describing them. The result was a big graph database with approximately 3,000
nodes and 10,000 weighted relationships. Over the global network we calculated
the graph measures to discover the relevant nodes in the graph.



5.3   Results


After computing all the algorithms, we focused our attention on the local network
of a user in particular. Results of centrality computation are presented in Table 2.
We can notice that higher centrality measure values allow us to suggest the most
relevant points of interest for this user. For example we can see that “Catedral
Puebla” is an element with high betweenness which means that it is a relevant
place in the city and that element should be recommended as “You can’t miss”.
Another relevant element is the “Museo Revolución” because it shows the highest
flow closeness. In the case of flow eccentricity, we can see the elements that are
far away from user preferences giving the opportunity to explore new things and
be different.
14

6    Conclusions

We have presented a first approach to a graph-based recommendation model that
takes advantage of social metrics and recommends points of interest to citizens
and/or visitors in a smart city. The proposed model expresses the semantics of
relationships that exist between users and points of interest through terms that
define a profile for the items. This new approach, using particularly flow cen-
tralities, considers semantic predominance of terms for defining and exploiting
the relationships among user profile preferences as well as the descriptive char-
acteristics of points of interest. Recommendations can then be extracted based
on the knowledge represented in the graph.
    In order to validate the recommendation model, the recommendation en-
gine was implemented and has shown that interesting recommendations could
be suggested to users, considering not only their preferences, but also taking
into account suggestions coming out from preferences of other members of the
social network related to them by the friendship relationship. The graph-based
recommendation model also proposes to explore points of interest that are very
different to user preferences, inviting him to explore new points of interest in the
city.
    The implementation of the recommendation engine is a challenging task be-
cause of the data volume and the complexity of required calculations to evaluate
flow centralities and semantic predominance. This challenge not only raised new
questions but also opened interesting opportunities for dealing with performance
issues. Preliminary results were presented, showing that the use of social met-
rics in any real recommendation system must include a specialized components
for solving distributed and concurrent processing tasks. Even though nowadays
there are advanced and efficient solutions for managing big data, adequate use
of graph-based solutions for modeling social networks still remains as the core
problem of a recommendation engine.
    The framework presented in this paper was tested through a prototype that
demonstrated the validity of our proposal. The recommendation engine is avail-
able through a REST APIs. These web services can be easily integrated into web
or mobile apps. Application domains include intelligent tourism, (as in those de-
scribed in this paper), as well as other areas of interest for citizens, such as
administrative services in a Smart City.


7    Future Work

The prototype will be extended and adapted to include specific information on
the cities of Puebla in Mexico and Shanghai in China. Also, different aspects
could still be improved in the recommendation engine to contribute to enrich
the user experience in a smart city:

 – Incorporation of new semantic filters to propose lists of objects of interest;
   proposing for instance only the points of interest in the proximity of the user’s
                                                                                     15

   location and considering the time when the user queries the recommendation
   system.
 – In absence of explicit evaluation of user preferences, we will explore the in-
   tegration of the Sentiment Analysis component developed by our group [12],
   to offer the possibility to add open comments and to evaluate automatically
   their polarity.
 – Routes recommendation: from the list of recommended points of interest,
   different alternative routes can be built. A prototype of a mobile application
   has been already developed for evaluating the interaction with users [16]. The
   prototype exploits data from Foursquare and recommends points of interest
   based on ratings made by users. The integration with the recommendation
   engine needs to be completed.

Acknowledgments Special thanks to our colleagues C. Thovex, F. Arámburo,
E. Castillo and D. Báez-López for their significant contributions in this work.
This project was partially supported by CONACYT-PROINNOVA No.198881,
CONACYT-OSEO No.192321, and China’s Foreign Experts Recruitment Pro-
gram.


References
 1. Anand, D., Mampilli, B.S.: Folksonomy-based fuzzy user profiling for improved
    recommendations. In: Expert Systems with Applications 41, 2424–2436 (2014)
 2. Bobadilla, J., Hernando, A., Ortega, F., Bernal, J.: A framework for collaborative
    filtering recommender systems. Expert Systems with Applications 38, 14609–14623
    (2011)
 3. Bobadilla, J., Ortega, F., Hernando, A., Bernal, J.: A Collaborative Filtering Ap-
    proach to Mitigate the New User Cold Start Problem. Know.-Based Syst. 26,
    225–238 (2012)
 4. Bobadilla, J., Ortega, F., Hernando, A., Gutiérrez, A.: Recommender systems sur-
    vey. In: Knowledge-Based Systems. ACM 46, 109-132 (2013)
 5. Cantador, I., Fernández, M., Vallet, D., Castells, P., Picault, J., Ribière, M.: A
    Multi-Purpose Ontology-Based Approach for Personalised Content Filtering and
    Retrieval, in: Wallace, D.M., Angelides, P.M.C., Mylonas, D.P. (Eds.), Advances
    in Semantic Media Adaptation and Personalization, Studies in Computational In-
    telligence. Springer Berlin Heidelberg, pp. 25–51 (2008)
 6. Cervantes, O., Thovex, C., Trichet, F.: Dynamic recommendations for smart cit-
    izens based on socio-semantic network analysis.: In: International Conference on
    Digital Intelligence. (2014)
 7. Chang, C.-C., Chu, K.-H.: A Recommender System Combining Social Networks
    for Tourist Attractions. In: Proceedings of the 2013 Fifth International Conference
    on Computational Intelligence, Communication Systems and Networks, CICSYN
    ’13. IEEE Computer Society. Washington, DC, USA, pp. 42–47 (2013)
 8. De Campos, L.M., Fernández-Luna, J.M., Huete, J.F., Rueda-Morales, M.A.:
    Combining content-based and collaborative recommendations: A hybrid approach
    based on Bayesian networks. International Journal of Approximate Reasoning 51,
    785–799 (2010)
16

 9. Le Merrer, E., Trédan, G.: Centralities: Capturing the Fuzzy Notion of Importance
    in Social Graphs, in: Proceedings of the Second ACM EuroSys Workshop on Social
    Network Systems, SNS ’09. ACM, New York, NY, USA, pp. 33–38 (2009)
10. Eyharabide, V., Amandi, A.: Ontology-based user profile learning. Appl Intell 36,
    857–869 (2011)
11. Freeman, L.C., Borgatti, S.P., White, D.R.: Centrality in valued graphs: A measure
    of betweenness based on network flow. Social Networks 13, 141–154 (1991)
12. Gutiérrez, E., Cervantes, O., Báez-López, D., Sánchez, J.A.: Sentiment Groups
    as Features of a Classification Model Using a Spanish Sentiment Lexicon: A Hy-
    brid Approach. In: Carrasco-Ochoa, J.A., Martı́nez-Trinidad, J.F., Sossa-Azuela,
    J.H., López, J.A.O., Famili, F. (Eds.), Pattern Recognition. LNCS, vol. 9116,
    pp.258–268. Springer International Publishing (2015)
13. Hwang, C.-S., Su, Y.-C., Tseng, K.-C.: Using Genetic Algorithms for Personalized
    Recommendation, in: Pan, J.-S., Chen, S.-M., Nguyen, N.T. (Eds.), Computational
    Collective Intelligence. Technologies and Applications. LNCS, vol. 6422, pp.104-
    112. Springer Berlin Heidelberg, (2010)
14. Movahedian, H., Khayyambashi, M.R.: Folksonomy-based User Interest and Disin-
    terest Profiling for Improved Recommendations: An Ontological Approach. J. Inf.
    Sci. 40, 594–610 (2014)
15. M. Newman.: A measure of betweenness centrality based on random walks. Social
    Networks, vol. 27, no. 1, pp. 39 – 54, (2005)
16. Pedraza, D.: RoX: Sistema de recomendación basado en preferencias. MSc Thesis.
    Universidad de las Américas Puebla (2015)
17. Robertson, S., Zaragoza, H.: The Probabilistic Relevance framework: BM25 and
    Beyond. Found. Trends Inf. Retr. 3, 333–389 (2009)
18. Saiph Savage, N., Baranski, M., Elva Chavez, N., Höllerer, T.: I’m feeling LoCo: A
    Location Based Context Aware Recommendation System.In: Gartner, G., Ortag,
    F. (Eds.), Advances in Location-Based Services. Springer Berlin Heidelberg, Berlin,
    Heidelberg, pp. 37–54 (2012)
19. Schiaffino, S., Amandi, A.: Intelligent User Profiling. In: Bramer, M. (Ed.), Ar-
    tificial Intelligence An International Perspective. LNCS, vol. 5640, pp.193–216.
    Springer Berlin Heidelberg, (2009)
20. Shinde, S.K., Kulkarni, U.: Hybrid personalized recommender system using
    centering-bunching based clustering algorithm. Expert Systems with Applications
    39, 1381–1387 (2012)
21. Thovex, C., Trichet, F.: Semantic social networks analysis. Soc. Netw. Anal. Min.
    3, 35–49 (2012)
22. Wang, H., Terrovitis, M., Mamoulis, N.: Location Recommendation in Location-
    based Social Networks Using User Check-in Data. In: Proceedings of the 21st ACM
    SIGSPATIAL International Conference on Advances in Geographic Information
    Systems, SIGSPATIAL’13. ACM, New York, NY, USA, pp. 374–383 (2013)
23. Ye, M., Yin, P., Lee, W.-C.: Location Recommendation for Location-based Social
    Networks, in: Proceedings of the 18th SIGSPATIAL International Conference on
    Advances in Geographic Information Systems, GIS ’10. ACM, New York, NY,
    USA, pp. 458–461 (2010)
24. Zanker, M., Jessenitschnig, M.: Case-studies on exploiting explicit customer re-
    quirements in recommender systems. User Model User-Adap Inter, Jour. Pers.
    Res., Springer 19, 133–166 (2009)

</pre>