PULPA DE REMOLACHA
5. CONCLUSIONES Y RECOMENDACIONES
The contemporary online social networks are diverse in nature and promote a wide variety of activities. The scientific literature suggests several classifications for online social networks depending on their functionality and features. Our main concern is to streamline personal information disclosure and address privacy issues in online social networks. Therefore, we present three classifications that distinguish OSNs from data- centric and social perspectives. The first classification is suggested by Ho [57] and the author based his classification on two criteria: how OSNs affect the privacy of their users and which type of personal content is exchanged among users. The author divided OSNs into four categories and for each category, some illustrative examples are provided.
2.2.3.1 Personal Online Social Networks
Online social networks of this category focus on providing the opportunity for users to connect with their family, friends, and acquaintances. Typical examples of personal online social networks include Facebook and Google+. The user profiles are one of
the core functionality of these OSNs. The profile contains huge amount of personal information
2.2.3.2 Professional Online Social Networks
The main purpose of this kind of social networks is to connect users with their busi- ness contacts and help them to find a job or look for employees. A typical example of professional online social networks is LinkedIn and Xing. The information avail- able on these OSNs is of professional nature which includes details about expertise, recommendations, and job offers.
2.2.3.3 Interest Oriented Online Social Networks
These OSNs allow users to share their hobbies and interests. Typical examples of this kind of OSNs are Last.fm and Flixster. The information shared on these services cannot be used to directly identify a user. The problem of privacy is not their core issue and these services are considered to be less sensitive with respect to user privacy.
2.2.3.4 Functionality Oriented Online Social Networks
These online social networks are known for their specific functionalities such as photo sharing, social bookmarking, and microblogging. Typical examples of these OSNs in- cludes Twitter, Flicker, Instagram and LiveJournal. These OSNs does not necessarily capture demographic information but rather a large amount of other personally iden- tifiable information such as photos. Another attempt to classify online social networks on the basis of pseudo-scientific literature was made by Beye et al. [61]. The authors classify suggest two broad categories on the basis of connections and content focused on online social networks.
2.2.3.5 Connection Oriented Online Social networks
Connection OSNs focus on the social connections and interactions between users, by providing users with a social contact list, channels for interaction, or matching ser- vices. Their general purpose is usually to connect users to new or existing friends and acquaintances or to provide an easy way to maintain such relationships.
Business: These OSNs aim to provide professionals with useful business contacts.
Searching for profiles does not always require signing up. Profiles display a users capabilities and work field as well as a means to contact that user. This is usually done through the OSN via messages. Users can also add other users to their network (connection) so that other professionals can see who the user is working or has contact with. An example of this class is LinkedIn.
Socializing: Fitting the more traditional view of social networks. Here users can
connect with current friends and find new ones. All types of information found in an OSN are also found in this class, often a lot of this information is public. The revenue for the OSN provider often comes from advertisements and selling information about the OSN, but can sometimes be combined with a subscription for additional functionalities (as with Hyves). In order to attract and keep users this type of OSN usually has a lot of additional functionalities such as social and competitive games. For a user the value of such an OSN is often largely determined by the number of friends on the OSN. Some well known examples of this class are Facebook, and Google+.
Dating: Dating sites are websites that aim to help users find the love of their life, many
of which incorporate OSN aspects these days. Each user has login credentials and usually a profile to attract potential lovers. Connections are typically in the form of love interests, but friendship links are also common; groups may also exist. Traversing the OSN is often based on searching or recommendations rather
than through navigating existing connections. Messages exchanged between users are often kept private to these users, although in some cases comment sections, viewable by connections, are offered. Example dating sites are match.com 3.
2.2.3.6 Content Oriented Online Social Networks
Content OSNs focus more on content provided by or linked to by users. This content can be multimedia or information like knowledge, advice, or news. The social interactions with other users usually revolve around and are driven by a search for information or the exchanging of said media.
Multimedia Content Sharing: Sharing of user-generated content can happen within
a selected group, such as friends or family, or a far wider audience. Content that is shared is usually multimedia; this is often of potential interest to a wide audi- ence, and even for selected audiences, e-mailing such content is cumbersome and often impossible due to size of the data. Uploading content generally requires users to sign up and log in; sometimes viewing content also requires logging in. The content tagging and recommendation may be an integral part of the system. Examples are Instagram4 and youtube5.
News Sharing: Some OSNs focus on world news or gossip, but a multitude of micro-
blogging OSNs provide a stage mainly for sharing personal news, opinions, and experiences. Examples are Twitter6.
Hobbies/Entertainmen: Many OSNs focus on audiences that have similar interests
and hobbies. Such OSNs may involve multimedia uploads, recommendation, or advice sharing elements, but the main distinguishing feature is their homogeneous 3 www.match.com 4www.instagram.com 5 www.youtube.com 6www.twitter.com
audience. This means that the topic of the OSN mainly determines its character and appeal for users. Examples are Xbox Live 7
The latest classification of online social networks from semantic web perspective is given by irfan et al. [62]. The authors categorize online social networks into three broad categories which include context based OSNs, content based OSNs and media based OSNs. The content-based online social networks allow the text-based interactions among users such as microblogging, social news, etc. The media based online social networks provide user interaction through various multimedia formats such as video and audio. The integration of semantic web technologies with online social networks can be more useful and productive for development of the intelligent social communica- tional services. According to authors, the past literature focused on content and media based OSNs, whereas most integral and crucial perspective related to social semantic was overlooked. The lack of semantic analysis was major barrier for effective intelli- gent social communication services. We will describe the context based online social networks in following section, as far as content and media based OSNs are concerned these already addressed in the previous section.
2.2.3.7 Context-based Online Social Networks
The context based online social networks provide an appropriate platform for the inte- gration of physical and logical contextual information that can be gleaned from various sources such as user profile, interaction and communication pattern of the users. An integration of the contextual information with interactive computing can be a promis- ing solution for the development of effective intelligent social communication services. There are three different type of context-based online social networks. The brief de- scription of these types is given below:
Social Semantic Web The social semantic web implements ontologies for context- based knowledge management.
Social Search The social search is shared effort of a group of users to obtain the rel-
evant information. The collaboration is an important aspect of an social search where multiple users can participate by suggesting different keywords, query syn- tax, and query reformation.
Social Recommendations The social recommendation target social media domain
and include online social relationships as an additional input parameter. There are two main categories of social recommendation systems which includes content- driven recommendation systems and collaborative-filtering based recommenda- tion systems.