Mining Social Website For Decision Support

Published Date: 02 Nov 2017

Abstract-Now a days social websites are very popular within the peoples. These sties like Facebook, twitter, G+ are gaining the popularity rapidly by forming a network. In various areas like healthcare, education, businesses, insurance these websites plays important role. Social network analysis uses different mining techniques for finding the required information which is useful for decision making.

This paper discusses the patient and doctor social websites to discuss about their problems and solutions in healthcare. Anal y sing by correlation, clustering and association for social media using insurance twitter posts .so how it is useful to take decisions for fulfilling the market requirements and customer services.

Keywords— data mining, social network analysis , social network mining, clustering, association analysis

What is a Social Network?:

A social network is defined as a set of actors (individuals) and the ties (relationships) among them. These relational ties and actors compose the fundamental interests in social networks. Also, a social group can be defined as a set of people who have common interests, i.e., like the same subjects, share their experience, express similar ways of thinking and reacting, share the same opinions, do similar things, and have the same goals. They actively exchange information. In the presence of new events, they discuss with each other to decide what to do.

Network analysis provides two purposes, revealing the underlying social structures and discovering the dynamic interactions among social actors. Network analysis identifies the system’s structure through examining the relations among the system components, its actors.

Social networking is an act of interacting/ sharing fun and some information popularly called profiles with known or unknown people (called friends) freely online. A social network service is an online interface, service, that enables users exchange information and relate socially. This consists of a representation of each user (often through a profile), his/her social links, photos, fun and a variety of additional services. Most social network services are web based and provide means for users to interact over he internet, such as email and internet messaging.

When we think of a social network, we think of Facebook, Google+, or another website that is called a "social network," and indeed this kind of network is representative of the broader class of networks called "social." The essential characteristics of a social network are:

1. There is a collection of entities that participate in the network. Typically, these entities are people, but they could be something else entirely.

2. There is at least one relationship between entities of the network. On Facebook or its like, this relationship is called friends. Sometimes the relationship is all-or-nothing; two people are either friends or they are not. However, in other examples of social networks, the relationship has a degree. This degree could be discrete; e.g., friends, family, acquaintances, or none as in Google+. It could be a real number; an example would be the fraction of the average day that two people spend talking to each other.

3. There is an assumption of non randomness or locality. This condition is the hardest to formalize, but the intuition is that relationships tend to cluster. That is, if entity A is related to both B and C, then there is a higher probability than average that B and C are related.

Social Networks as Graphs: Social networks are naturally modeled as undirected graphs. The entities are the nodes, and an edge connects two nodes if the nodes are related by the relationship that characterizes the network. If there is a degree associated with the relationship, this degree is represented by labeling the edges.

Example : Figure 1 is an example of a tiny social network. The entities are the nodes A through G. The relationship, which we might think of as "friends," is represented by the edges. For instance, B is friends with A, C, and D.

Social network mining:

In social network mining, we apply data mining algorithms to study large-scale social networks. Social network mining has attracted a lot of attention for many reasons. For example, studying large social networks allows us to understand social behaviors in different contexts. In addition, by analyzing the roles of the people involved in the network, we can understand how information and opinions spread within the network, and who are the most influential people (Fig. 1). In addition, since social network users may receive too much information from time to time, social network mining can be used to support them by providing recommendations and filtering information on their behalf. [1].

In social network mining, we generally ask three broad questions:

(1) What are the characteristics of the social network?

(2) How can we model the network?

(3) How can we support its users?

When trying to answer the first question, we aim to identify different properties of a given social network.

For example, what do people do in this social network?

Do they exchange messages, or do they share items among themselves? We can also ask, for any two persons, what is the probability distribution of the distance between them? Are there any clusters or communities within the network? Answering these questions enables us to understand how information flows and how social relations in the network evolve. For example, some research on trying to understand the social networks of Twitter and Flickr has been done.

After understanding the characteristics of a particular social network, we may want to construct a mathematical model that explains the processes in the network. A mathematical model lets us predict future changes in the network. For example, what is the probability of a new edge between two given persons?

their social circles?

The above three steps form a cycle (Fig. 3) that one can travel along in order to continuously gain more and more insight into how people interact and then improve the experience of the social network’s users, thus attracting more people to participate in it.

There are two types of social networking

Mobile Social Network Mining

Web based Social Network Mining

Now we will see some areas where data mining is used for social wesites.

Managing Healthcare Through Social Networks

Healthcare social networks can be either physician or patient oriented. Physician social networks provide an online technical infrastructure for doctors to share clinical cases, images, videos, and medical knowledge. Patient social networks emphasize direct patient support, promoting disease awareness, and positive and proac­tive behaviors to stay healthy while living with a disease. Both network types can greatly enhance healthcare management.

Physician networks :The large number of physicians attracted by physi­cian social networks is surprising. Manhattan Research, a global pharmaceutical and healthcare market research company, found that 60 percent of US physicians use or plan to use a physician social network.6 These platforms bring thousands of physicians together to exchange the latest in medical advances. Sermo (www.sermo.com) is the largest online physician community, with more than 112,000 members representing 68 specialties as of March 2010; members’ average age is 47, and their average ex­perience level is 13 years. Other popular physician social networks such as Ozmosis (www.ozmosis.com) and Doc­torNetworking (www.doctornetworking.com) reported anywhere from 3,000 to 10,000 physician members each as of September 2009.Members seem largely satisfied with the networks.

Benefits:

In a physician social network, physicians exchange views about drugs, devices, and treatment options and can use their knowledge from daily prac­tice to ask and answer specific clinical questions that are not obvious in the medical literature. physi­cians can better solve problems, collaborate on difficult cases, and predict future events through a network than they could individually or even in a small group. physician social networks contribute to the emergence of trends and insights in medications, devices, and treatment. Physicians can discuss ongoing research and help speed the process of bringing advances to pa­tient care.

Patient networks Benefits:-

Patient social networks support e-patients worldwide by promoting disease awareness and positive and proactive behaviors to stay healthy while coping with disease. E-patients or e-caregivers—friends or family members—can access content, connect with others, or collaborate with others in exploring treatment options or other concerns.9 PatientsLikeMe is a patient network with more than 58,000 registered patients as of March 2010.

Concerns:-

As in physician social networks, the biggest concern is trust. To what extent can people trust the con­tent of patient social networks? Most medical advice and comments come from users, and sites do not verify their validity before publication Privacy is also a concern. PatientsLikeMe warns individ­uals before they register that their personal records will be visible to others and that drug or insurance companies can pay for access to the aggregated information.

In healthcare data mining are intended to be helpful tools that can improve the physicians’ performance and make the diagnosis process more objective and more reliable. Diagnosis rules may be automatically derived by means of clustering, machine learning, and association rules from the descriptions of the patients treated in the past for which the final diagnosis were verified.

Data Mining Applied to Insurance Twitter Posts

Objective:

how data mining and text analytics can be applied to social media to identify key themes in the data. the analysis of Twitter posts related to the keyword Allstate. Allstate was chosen purely based on the public availability of historical Twitter data. to provide a detailed analysis of Twitter activity related to Allstate, but to demonstrate how analytics can specifically be applied to social media information related to a property and casualty insurance company.

B. Data Processing:

The source of the data for this analysis is twapperkeeper.com, which is a service that captures and archives Twitter posts. There are third party applications which can capture social media data from websites, and these applications appear to be both web-based services as well as stand-alone programs. Developers can also create computer programs which monitor and capture information from their social media sites. Programs also scrape information from screens, and this can be applied to monitoring and collecting social exist which media data from websites. Ultimately, companies will need to work with their information technology

departments to determine the best approach for collecting and storing social media data. The first step in analysing the text data is to remove all the punctuation and symbols. This Once the tweet has been cleansed of punctuation marks and symbols, then the tweet can be parsed into words. This parsing occurs by the purpose of the analysis of social media is to identify patterns occurrence of specific words together. and trends that are present in the information which may be of further use to the insurance company. To achieve this goal, we need to identify patterns and combinations of words that will indicate themes and ideas. One step in doing this is a simple correlation analysis, which will identify correlations between pairs of words. There are also two additional types of analyses that will be performed on the data. The first will be a clustering analysis which will group tweets based on their similarities or dissimilarities. The second will be an Association Analysis which analyses the identifying spaces and using these spaces as the indication of one word ending and another word beginning. The purpose of the analysis of social media is to identify patterns and trends that are present in the information which may be of further use to the insurance company. To achieve this goal, we need to identify patterns and combinations of words that will indicate themes and ideas. One step in doing this is a simple correlation analysis, which will identify correlations between pairs of words. There are also two additional types of analyses that will be performed on the data. The first will be a clustering analysis which will group tweets based on their similarities or dissimilarities. The second will be an Association Analysis which analyzes the occurrence of specific words together.

C. Challenges With Social Media Analysis:

One of the first challenges will actually be accessing and collecting this information., there are applications available which will allow companies to begin collecting and analyzing social media data, and companies may also have the ability to build internal programs that do this. The key is to make sure that the data is being collected in a consistent and complete matter, and that it is easily accessible for analysis. There are also challenges in analyzing social media as it relates to analyzing text data. One of these challenges relates to the context. Another challenge in terms of the analysis of social media data is understanding customer sentiment. Words in a tweet are simply that, words. They do not carry the normal emotion that is present in a face-to-face conversation, in which case the listener could detect happiness, sadness, sarcasm, etc. There are things that users attempt to do to try and convey different emotions (smiley face, sad face, "lol," TYPING IN ALL CAPS, etc.), but even when a person reads an electronic communication they might get the sentiment wrong. Next social media data is unfiltered. There are no system edits that ensure the social media data that was captured is accurate, and this may result in false information and statements that are driven by pure emotion rather than fact. Lastly the world of social media is not limited to those that use the English language.

D. Applications:

Provide customer service

to better understanding of customer

sentiment about the company.

to identify broader trends in the market that the company may be able to take advantage of.

The Future Of The Social Web:

Social networks represent the digital reflection of what humans do: We connect and share. While brands naturally want to get in on this furor of activity, there’s a big problem — the social information about people, their profiles, and their friends is locked up in separate networks, frustrating both the consumers who use them and the brands who want to connect with them. But the Social Web is about to evolve into something much broader than a few social network sites: a consistent backdrop for every online activity. Portable social IDs and the changes they enable will transform how consumers, brands, and social networks interact. is online social experience will evolve through five eras . era of social relationships: is was the first stage of the Social Web, starting in the mid- 1990s with communities like AOL and maturing a few years ago. In this era, people connected to each other using simple profiles and friending features to share information, discussions, and media.1 While this era is the foundation of the changes to come, in this document we’ll be concentrating on the next four eras — the future of the Social Web.

1. era of social relationships:. is was the first stage of the Social Web, starting in the mid-1990s with communities like AOL and maturing a few years ago. In this era, people connected

to each other using simple profiles and friending features to share information, discussions, and media.1 While this era is the foundation of the changes to come, in this document we’ll be concentrating on the next four eras — the future of the Social Web. [4]

2. era of social functionality: Today’s social networks have evolved beyond "friending" into platforms that support social interactive applications and provide new meaning and utility to communities. Even so, social relationships are still locked up within sites.

3. era of social colonization. In the next stage of social evolution, starting later in 2009,

technologies like OpenID and Facebook Connect will let individuals traverse the Internet with their social connections along for the ride. the boundaries of social networks and traditional sites will blur, making every Web site into a social experience.

4. era of social context: Next year, as sites begin to recognize people’s personal identities and their social relationships, they will customize visitors’ experiences based on their preferences, their behaviors, and who their friends are. In addition to enabling more intense social applications, in this stage social networks will absorb features of email and become a base of operations for everyone’s online experiences.

5.era of social commerce. Starting about two years from now, as social networks become the repository for identities and relationships, they will become more powerful than corporate Web sites and CRM systems. Communities will become the driving force for innovation. As a result, brands will cater to communities, resulting in a power shin toward the connected customer.

Conclusion

As explained above data mining is very useful tool for social websites to finding useful patterns in social site data so the appropriate decisions can be made.