Web Mining and Social Multimedia - Research Paper Example

? Web Mining and Social Multimedia Other (s) Web mining is the use of data mining techniques to extract data from web documents. Data mining allows for analysis of data in order to make rational decision based on the data report collected. When such data is being analyzed based on the web documents, especially with a wide spread of social multimedia, the information gathered can benefit research institutions, businesses and economy overall. However, how does data mining filters the information gathered from the web from undesirable and inaccurate data that is so often giving us difficulties in gathering valuable and high-quality results? It is common knowledge that the data found on the web is unstructured, dynamic, complex and huge in amount. This results in difficulties for analyzing such data. What techniques and applications are used in Web Mining to overcome these difficulties? The answers to these questions could benefit both research and industry communities. Web Mining and Social Multimedia Introduction Web mining refers to the application of data mining techniques to extract patterns from the web. Generally data mining allows for analysis of data in order to make rational decision based on the data report collected. It is common knowledge that the data found on the web is unstructured, dynamic, complex and huge in amount which results in difficulties for analyzing such data. Consequently the information gathered by web mining can be further evaluated using various software or through the traditional data mining parameters such as classification, clustering and association. There are three main axes of web mining which include content mining, usage mining and structure mining. Content mining is usually applied in the examination of data collected by web spiders and search engines. On the other hand, structure mining is used when examining the structure of given websites while usage mining is generally used to study data related to user’s browser as well as the data collected by the forms users usually submit during their web transactions. When such data is being analyzed based on the web documents, especially with a wide spread of social multimedia, the information gathered can benefit research institutions, businesses and economy overall. Additionally web mining can potentially be used in customer relationship management by helping to evaluate the customer behavior, effectiveness of the website as well as quantify the success of the marketing campaign used in the World Wide Web. Although web mining has numerous potential benefits particularly with regard to the interpretation of meaningful data, the technology has also been regarded as a disruptive technology due to some of the risks it poses both to personal and cooperate privacy (Domingos, 58). For example, the sophisticated technologies used in web mining have significantly increased the risk of information abuse as well as privacy violation. As many social media platforms continue to urge their users to become more transparent by revealing their personal information, the privacy of such users may be compromised. There are a number of data base technologies through which web mining can be used to discover the patterns in data. Some of the commonly used database mining techniques used in web mining include clustering, association and data classification. The difficulties in gathering quality data using web mining techniques usually arise from the fact that there are currently no agreed upon quality assessment models as well as the difficultly that arises from handling the quality of information particularly during the query processing and integration of data. In web mining, some of the scenarios in which the problem of data quality may arise include during the integration of scientific or business data and during the dissemination of the collected data. History of web mining The concept of web mining has rapidly grown in a short period of time both in terms of practitioner communities and in research. The World Wide Web was invented in 1989 by CERN Company and in 1990; the first browser and server came into use. The rapid spread of the web was particularly facilitated by the Mosaic browser was developed by Eric Brina and Marc Andreessen in 1993. The World Wide Web basically serves as the primary platform on which web mining is conducted. On the other hand, web mining is a well- organized and established field with many applications over the last many years experiencing a vast and rapidly growing amount of multimedia content that has become available online in the internet. The term web mining was first used in 1996 to refer to the application of various data mining techniques in the extraction of knowledge from web data sources in a task oriented manner. The initial approach used in data mining was process centric and web mining was largely defined as a sequence of tasks and processes. In 1997, a more data oriented approach was adopted in the definition of web mining. The data centric view of web mining increasingly became widely accepted and has been adopted in many papers and journals. Continuous rapid growth of the World Wide Web has imposed new methods of web mining that can effectively help in the processing of the increasingly huge amounts of data and the diverse formats in which the information is currently published. Some of the problems that have faced web mining include how to effectively detect relevant information as well as address the needs of individual users. Web mining features with growth of social multimedia The emergence and rapid proliferation of various social net work platforms such as Twitter, Facebook, Linkedin and other multimedia network sites has been one of the significant events witnessed in the last few decades. These network sites have not only increasingly become popular but they have also form part of the daily lives of millions of people across the globe. Social network sites are usually rich in content and generally have huge amounts of multimedia content that requires mining and analysis (Naamarn, 6). There are a number of web mining features that can potentially be used in the analysis of the huge amounts of multimedia data in the social network sites. Social networks have greatly and dramatically influenced the growing amount of multimedia content due to the fact that those who use it become active and important producers, distributors and consumers of such multimedia context. Web mining techniques and its unproven statements in it need to get expanded in order to include new arriving trends such as heterogeneous new web data. This has conceptualized and introduced the concept of social multimedia mining as a new emerging research area that puts together web mining research activities, multimedia research and social media research. Multimedia data mining is generally defined as the process of extracting the patterns of media data such as images, video, audio and texts that are otherwise not accessible to the normal queries and associated contents. Apart from the extraction of multimedia data, social data mining also involves the discovery of communities, personalization and a number of search methods for social activities such as finding friends. In social media networks, web mining involves extracting useful information from survey logs by user’s history. It is the process of finding out what the users are looking for on the internet. Some users may be looking at contextual data while others may be interested in multimedia data which is of great importance in research services. According to Lappas (18), multimedia data mining techniques can effectively be used to enable individuals to share their knowledge, opinions as well as benefit from each other’s experience. For example, this can be achieved by mining and redistribution of information obtained from computational records of social activity such as system usage, usage messages and hyperlinks among others. Web mining has been highly applied in social multimedia in areas like understanding customer behavior to evaluate effectiveness of a particular website and help quantify the success of a marketing campaign. It has also been used in form of structure mining to examine data related to the particular website’s structure while usage mining examines data related to a particular user’s browser as well as data collected by forms the users may have submitted during web transactions in social; multimedia. Information through web mining is therefore evaluated by software graphing using traditional data mining which is an important aspect in the social multimedia. Data mining filters the information gathered from the web from undesirable and inaccurate data that is so often giving us difficulties in gathering valuable and high-quality results (Ernandez, 115). Generally the mining of social network data is an important field that has numerous application areas. For example, in law enforcement, web mining of social network data can be used to track criminals, their activities and connections. In commerce, web mining is increasingly being adopted to help understand the relationships and interactions between the business organizations and their customers. For example, online stores such as Amazon.com have adopted an approach of personalized customer experience using a host of web mining techniques such as click path analysis of the users and analysis the association between pages visited by the customers. Additionally the knowledge gained from web mining has been used as a key intelligence in the development of features such as instant recommendations, wish lists and other features that are currently being used by many online multimedia platforms. Although web mining has potential immense benefits in the social multimedia networks, its use has however come with a number of challenges and increasing risks of privacy violations. Generally web mining in social multimedia sites has also posed a number of risks privacy of individual web users. According to Cadez (281), the sophisticated technologies used in web mining have significantly increased the risk of information abuse as well as privacy violation. As many social media platforms continue to urge their users to become more transparent by revealing their personal information, the privacy of such users may be compromised. Web mining also poses a significant threat to the ethical values regarding the individuality and privacy of persons. This is particularly because the techniques used in web mining usually make it difficult for individuals to autonomously control the dissemination and unveiling of data about their private life. For instance, when data is published in the social media network using a particular context is mined and combined with other data in different contexts, when content and structure mining can pose a serious threat to the security of such data. On the other hand, web usage mining raises the vulnerability of private information when the web and social network users are traced and their actions analyzed without their consent or knowledge. There is however a number of solutions that have been developed to address some of the privacy related problems associated with web mining. For example, algorithms have been developed to modify and process the original private data to ensure that they remain private even after the web mining process. Some of the other potential solutions to the privacy issue include data modification such as noise injection, rule hiding and data distribution. Real life applications of web mining in social multimedia There is a rapidly increasing interest of web mining some of the contemporary popular social network sites such as the Facebook (Foot, 14).This is particularly with regard to the important role that Facebook and other social media networks have increasingly played in the last decade. Facebook is increasingly being organized in a way that makes it easy to monitor the activities and preferences of its millions of users. With nearly 900 million users, Facebook is currently one of the largest social network sites with a huge user data including personal photos, private conversations as well as recording of travels, marriages and user preferences. Web mining in Facebook has taken a diverse number of forms. For example, Facebook users are usually prompted to fill out their profiles with personal details such as their age, email address and gender. Some people can also additional private details which include mobile phone number, relationship status among others. Additionally the latest Facebook redesign has introduced the use of profile pages known as timelines that prompt people to add their historical information such as the places they have worked or lived in. The messages and photos shared in Facebook are usually tagged with precise locations and Facebook has also begun tracking the activity of its users on other places outside Facebook. This has particularly been achieved by the use of “Like” button which usually appears on a number of websites and applications outside Facebook and the interests of users in particular brands, products or pieces of digital content can therefore effectively be tracked. Generally the information obtained through web mining techniques has been used in a number of areas in Facebook. For instance, the huge personal information collected from Facebook users has enabled the social network to identify and suggest friends for the users. Face book has also been able to monitor the basic motivations and patterns of human behavior of its users and some of the results have been published in numerous academic journals (Coyle, 101). Lastly, some of the information collected through web mining of Facebook has been sold to numerous ads and used for commercial purposes. The future applications of Web mining With the current rapid growth in information technology as well as internet accessibility, it has increasingly become difficult to analyze and extract relevant information from the huge databases. In the future a number of web mining techniques are likely to be to be applied to address some of these challenges (Berendt, 697). New technologies, communication and data strategies present an exciting opportunity for the future and social sciences both quantitative and qualitative to harness the power of the web. For example, currently web mining uses Facebook data to gauge economic effects during hard times on attitudes towards migration. Other web mining techniques will allow developers to access large data streams such as those controlled by Amazon or Facebook and then reconstruct them for various reasons that don’t follow the original data. Data reconstruction often combines data to create an aggregation, visualization or novel application from two or more data sources. This is likely to include the use of data modification techniques such as noise injection, rule hiding and data distribution Censorship efforts target online social networking communities across the globe. For example, sometimes certain countries may be barred to access major websites including details including texts images or videos. Lastly to solve the problem of privacy violation, the future web mining techniques are likely to adopt the use of algorithms that will enable social media developers to modify and process the original private data to ensure that they remain private even after the web mining process. Works Cited Berendt, Bryson K. (2012). More than modeling and hiding: of Web mining and towards a comprehensive view Data Mining and Knowledge Discovery. Privacy,24.3(2012), 697-737.Print. Cadez, D. Heckerman. Visualization of navigation patterns on a Web site using model based clustering. Knowledge Discovery and Data Mining, 2000. Web.10Nov.2012. Coyle, M. Freyne. Social information access for the rest of us: an exploration of socialYouTube. Proceedings of the 5th International Conference on Adaptive Hypermedia and Adaptive Web-Based Systems.8.3.(2008) 93-102.Print. Domingos, Richardson M. Mining the network value of customers, Knowledge Discovery and Data Minining. (2001):57-66.Print. Foot, K., Schneider. Analyzing Linking Practices: Candidate Sites in the 2002 U.S. Electoral Web Sphere. Journal of Mediated Communication, 8.4(2003): 14-28.Print. Hernandez, J. Ochoa. Biometrics in online assessments: A Study Case in High School Students, Conielecomp (2008):111-116.Print. Lappas, Gregory. From Web Mining to Social Multimedia Mining.2011 International Conference on Advances in Social Networks Analysis and Mining. Web.10Nov.2012. Naamarn, Mark, K..Social Multimedia: highlighting opportunities for search and mining of multimedia data in social media applications. Multimedia Tools and Applications.2010. Web.10Nov.2012. Read More