data that would be useful for a network anal ysis. This chapter presents a set of criteria for qualifying a system as a WbSn and another set for determining when information can be considered part of a relationship. Those principles guided an exhaustive survey of existing WBSNs followed by a discussion of trends in social network data sharing on the Semantic Web 2.2 Previous work This survey is motivated by the large body of work in social network anal ysis and in the study of online communities. While it would be impossible to cite all of the influential work related to social network anal ysis, the range of interest in the topic across nearly every academic field is impressive. These are just a few examples to give a sense of that Much of the foundational work in the analysis of social networks, and the major advances in the 20 century have been carried out in the fields of sociology, psychology, and communication(Barnes, 1972), Wellman, 1982), (Wasserman Faust, 1994). with a goal of understanding the function of relationships in social networks, and how they affect the social systems in which the networks exist, the research has been both theoretical and applied. Labor markets(Montgomery, 1991), public health( Cattell 2001), and psychology (Pagel, et al., 1987)are just a few of the spaces where social network analysis has yielded interesting results In the last five to ten years, a new interest has developed in the structure and dynamics of social networks to complement the work already being done in social network theory. Though one of the first, and most popular papers in this area- Milgram's Six Degrees of Separation"study(1967)-was conducted by a social scientist, the topic
11 data that would be useful for a network analysis. This chapter presents a set of criteria for qualifying a system as a WBSN and another set for determining when information can be considered part of a relationship. Those principles guided an exhaustive survey of existing WBSNs followed by a discussion of trends in social network data sharing on the Semantic Web. 2.2 Previous Work This survey is motivated by the large body of work in social network analysis and in the study of online communities. While it would be impossible to cite all of the influential work related to social network analysis, the range of interest in the topic across nearly every academic field is impressive. These are just a few examples to give a sense of that scope. Much of the foundational work in the analysis of social networks, and the major advances in the 20th century have been carried out in the fields of sociology, psychology, and communication (Barnes, 1972),(Wellman, 1982),(Wasserman & Faust, 1994). With a goal of understanding the function of relationships in social networks, and how they affect the social systems in which the networks exist, the research has been both theoretical and applied. Labor markets (Montgomery, 1991), public health (Cattell, 2001), and psychology (Pagel, et al., 1987) are just a few of the spaces where social network analysis has yielded interesting results. In the last five to ten years, a new interest has developed in the structure and dynamics of social networks to complement the work already being done in social network theory. Though one of the first, and most popular papers in this area – Milgram's "Six Degrees of Separation" study (1967) – was conducted by a social scientist, the topic
is of increasing interest to physical scientists. Their studies have addressed issues such as mathematical analyses of the structure of small world networks (Watts, 1999) community structure(Girvan, Newman, 2002), and how social network structure affects the spread of disease(Dezo, et al., 2002), (Jones, et al. 2003),(Newman, 2002) As the web emerged, online communities and social networks supported by the internet became a source of interesting data. Garton, et al.(1997) presented an excellent introduction to how traditional methods of social network analysis could be applied to these online communities. Work in this space was also embraced by the interdisciplinary field of human-computer interaction, which produced interesting work on designing and supporting online communities(Preece, 2000), their application to problems such as collaborative filtering(Kautz, et al. 1997) and electronic commerce Jung, Lee, 2000) The promise of social networks on the web is that they offer new opportunities to researchers across the board. with network topologies that can be automatically extracted from the web, the social networks provide a new, large source of data for the more mathematical and structural types of analysis. At the same time, users are participating in rich social environments online while building these networks. That holds promise for scientists interested in the general function of social interactions and because the contexts of these social networks is often very restricted(e. g. business networking among Asian-Americans), they can serve as a window into specific communities 2.3 Definitions There are many ways in which social networks can be automatically derived on the web: users connected through transactions in online auctions users who post within the same thread on a news group or message board, or even members of groups listed in an
12 is of increasing interest to physical scientists. Their studies have addressed issues such as mathematical analyses of the structure of small world networks (Watts, 1999), community structure (Girvan, Newman, 2002), and how social network structure affects the spread of disease (Dezo, et al., 2002), (Jones, et al. 2003), (Newman, 2002). As the web emerged, online communities and social networks supported by the internet became a source of interesting data. Garton, et al. (1997) presented an excellent introduction to how traditional methods of social network analysis could be applied to these online communities. Work in this space was also embraced by the interdisciplinary field of human-computer interaction, which produced interesting work on designing and supporting online communities (Preece, 2000), their application to problems such as collaborative filtering (Kautz, et al., 1997) and electronic commerce (Jung, Lee, 2000). The promise of social networks on the web is that they offer new opportunities to researchers across the board. With network topologies that can be automatically extracted from the web, the social networks provide a new, large source of data for the more mathematical and structural types of analysis. At the same time, users are participating in rich social environments online while building these networks. That holds promise for scientists interested in the general function of social interactions, and because the contexts of these social networks is often very restricted (e.g. business networking among Asian-Americans), they can serve as a window into specific communities 2.3 Definitions There are many ways in which social networks can be automatically derived on the web: users connected through transactions in online auctions, users who post within the same thread on a news group or message board, or even members of groups listed in an
HTML document can be turned into a social network. Many online communities claim t e or support social networks, but lack some of the properties one may expect of a social network. This work uses a very specific definition. A web-based social network must meet the following criterias 1. It is accessible over the web with a web browser this excludes networks where users would need to download special software in order to participate and social networks based on other technologies such as mobile devices 2. Users must explicitly state their relationship with other people qua stating a relationship. Although social networks can be built from many different interactions a Wbsn is more than just a potential source of social network data; it is a website framework that has the development of an explicit social network as a goal. This criteria rules out building social networks from auction transactions, co-postings, or similar events that link people when a connection is created as a side effect of another process. 3. The system must have explicit built-in support for users making these connections The system should be specifically designed to support social network connections This means that a group of friends who each maintain a simple HTML page with a ist of his or her friends would not qualify as a wbsn because HTML itself does not have explicit built-in support for making social connections. There must be some greater over-arching and unifying structure that connects the data and regulates how it Is presented and formatted 4. Relationships must be visible and browsable. The data does not necessarily have to be ublic (i.e. visible by anyone on the web) but should be accessible to at least the
13 HTML document can be turned into a social network. Many online communities claim to be or support social networks, but lack some of the properties one may expect of a social network. This work uses a very specific definition. A web-based social network must meet the following criteria: 1. It is accessible over the web with a web browser. This excludes networks where users would need to download special software in order to participate and social networks based on other technologies, such as mobile devices. 2. Users must explicitly state their relationship with other people qua stating a relationship. Although social networks can be built from many different interactions, a WBSN is more than just a potential source of social network data; it is a website or framework that has the development of an explicit social network as a goal. This criteria rules out building social networks from auction transactions, co-postings, or similar events that link people when a connection is created as a side effect of another process. 3. The system must have explicit built-in support for users making these connections. The system should be specifically designed to support social network connections. This means that a group of friends who each maintain a simple HTML page with a list of his or her friends would not qualify as a WBSN because HTML itself does not have explicit built-in support for making social connections. There must be some greater over-arching and unifying structure that connects the data and regulates how it is presented and formatted. 4. Relationships must be visible and browsable. The data does not necessarily have to be public (i.e. visible by anyone on the web) but should be accessible to at least the
registered users of a system. Websites where users maintain completely closed lists of contacts are not interesting for their social networking properties-neither to users or people performing a network analysis-and are thus ignored for these purposes. For example, some websites allow users to bookmark the profiles of other users and others allow users to maintain address books. Even when these lists are explicit expressions of social connections, they would not qualify a system as a WBsn if they cannot be seen and browsed by other users. One important note here is that the system itself does not need built-in browsing support. Rather, each user's data must be made accessible with unambiguous pointers to each social connection These criteria qualify most of the major social networking websites like Tickle. Friendster, Orkut, and LinkedIn while ruling out many dating sites, like Match. com, and other online communities that connect users, such as Craigs list or meetUp. com. Sites hat require users to pay for membership are included as long as they meet the criteria above Within these social networks, users are often able to say more about their relationships than simply stating they exist. However, it is easy to confuse functionality of a WBSN with actual information about a relationship. Again, it is helpful to have a set of criteria that establish when an action or datum qualifies as information about a elationship in the social network. 1.a basic social networking connection between individuals must exist before additional information can be added sites that allow users to rate others such as rating someone's appearance, often do not require that users have a connection
14 registered users of a system. Websites where users maintain completely closed lists of contacts are not interesting for their social networking properties – neither to users or people performing a network analysis – and are thus ignored for these purposes. For example, some websites allow users to bookmark the profiles of other users and others allow users to maintain address books. Even when these lists are explicit expressions of social connections, they would not qualify a system as a WBSN if they cannot be seen and browsed by other users. One important note here is that the system itself does not need built-in browsing support. Rather, each user's data must be made accessible with unambiguous pointers to each social connection. These criteria qualify most of the major social networking websites like Tickle, Friendster, Orkut, and LinkedIn while ruling out many dating sites, like Match.com, and other online communities that connect users, such as Craig's List or MeetUp.com. Sites that require users to pay for membership are included as long as they meet the criteria above. Within these social networks, users are often able to say more about their relationships than simply stating they exist. However, it is easy to confuse functionality of a WBSN with actual information about a relationship. Again, it is helpful to have a set of criteria that establish when an action or datum qualifies as information about a relationship in the social network. 1. A basic social networking connection between individuals must exist before additional information can be added. Sites that allow users to rate others, such as rating someone's appearance, often do not require that users have a connection –
anyone can rate anyone else. In order to be used as additional information about relationship, there must be a relationship between people in the first place. Thus, simple rating systems that do not require users to be socially connected are not counted 2. The information must be persistent. Many websites allow users to send messages or mini-messages(such as"winks"orsmiles"on dating-related sites). Since these are sent and do not persist as a label on the relationship, they are not a piece of information about a relationship. On the other hand, comments or testimonials about a person do persist on the website and are considered as free text descriptions of a relationshi 3. The information must be visible and modifiable by the user who added it. At the same time, the information does not have to be publicly visible. Some data, like trust ratings, are personal and users would not want this shared with others 2.4 A Survey of web-based Social Networks The goal of this survey was to profile every social network on the web that met he criteria above. The number of users and primary purpose of each website, along with what additional relationship information they support, if any, was gathered from each website Winks"or"smiles"are usually sent by clicking an icon on a member's page. That member then receives a small message letting them know that someone sent them a"wink"or a"smile These mimic their real world counterparts in that they are small indications of interest without requiring the sender to say much
15 anyone can rate anyone else. In order to be used as additional information about a relationship, there must be a relationship between people in the first place. Thus, simple rating systems that do not require users to be socially connected are not counted. 2. The information must be persistent. Many websites allow users to send messages or mini-messages (such as "winks" or "smiles"1 on dating-related sites). Since these are sent and do not persist as a label on the relationship, they are not a piece of information about a relationship. On the other hand, comments or testimonials about a person do persist on the website and are considered as free text descriptions of a relationship. 3. The information must be visible and modifiable by the user who added it. At the same time, the information does not have to be publicly visible. Some data, like trust ratings, are personal and users would not want this shared with others. 2.4 A Survey of Web-based Social Networks The goal of this survey was to profile every social network on the web that met the criteria above. The number of users and primary purpose of each website, along with what additional relationship information they support, if any, was gathered from each website. 1 "Winks" or "smiles" are usually sent by clicking an icon on a member's page. That member then receives a small message letting them know that someone sent them a "wink" or a "smile". These mimic their realworld counterparts in that they are small indications of interest without requiring the sender to say much