Big Data and its Role in Social Media: A study of Facebook
The concept of big data has gained application with the increase in the need for data transfer and storage in the application of conventional information technologies over time. Traditionally, databases performed roles that are currently associated with this concept. The evolution of data has been through advancement in terms of data characteristics and applicability, resulting in incremental innovation as the needs associated with data use also revolves (Madden 2012). For instance, big data has been described as a disruptive innovation based on its features and the benefits that have been associated with it (Lycett 2013). Conventionally, any data is characterized by volume, velocity, and variety. The distinction between the different types of data and the sole driver for the innovation from databases to big data is based on the quantity or level associated with the data features (Madden 2012).
While databases handled data in terms of gigabytes and megabytes, Big Data handles data in terabytes to Petabytes. This is an indication of volume expansion from databases to Big Data. Moreover, the velocity associated with big data depends on repetitive observations or decisions constantly being made. This implies that while databases and small data are associated with more intermittent data access, handling and storage, big data deals with continuous and highly dynamic data. In terms of variety, the databases and small data are restricted in terms of variety while big data deals with a range of data varieties (Zikopoulos et al 2012). It is because of these differences that there is increasing shift from the conventional methods to operations based on big data. The application of big data has transcended the confines of structure and limitations such as exhaustibility and data resolution. The move from the traditional storage of structured data to the present digital storage of unstructured data could not be achieved without big data (Cote 2014). Currently, big data find application in a wide variety of fields, from the government operations, the stock market operations, and other financial aspects, education, social media and scientific sectors. The benefits associated with big data are similarly diverse, ranging from improved cost effectiveness and innovation (Vis 2013).
Despite the wide applications, the roles played by big data on various fronts have not been studied extensively. The benefits of big data on some sectors have been looked into by some studies, although to a very narrow level. This clearly indicates that there is a narrow dearth of knowledge regarding the particular roles played by big data on these fronts. The application of big data on social media is thus a justifiable topic of study, as it will help to expand the information available about the roles played by big data in the communications sector. Facebook will be used as a case study in obtaining the required information. In carrying out the study, the research will be based on a literature review and an empirical analysis, which will be aimed at:
- Finding out the features that make big data applicable in various sectors
- Determining the scope of big data applications in social media
- Finding the roles played by big data in the operation of Facebook
Big Data has been associated immensely with the concept of business intelligence both in academia and in the business sector. According to Lycett (2013), big data is the foundation of business intelligence since the greatest resource for intelligent business operation is data. The need for intelligent data operation drives innovation and productivity in all business quarters. Although the need for big data is critical in all businesses, the level of need differs from one business to the other depending on the volume and variety of data that needs to be applied and managed. As in other business sectors, the social media has found immense application of the big data concept due to the large volume of data that is required to be handled. It may even be said that social media is the greatest applicant of big data. The application of big data and benefiting from its many advantages is however hindered by lack of comprehension of its scope and what it entails.
Non-human assemblages in Big Data usage
Traditionally, the characteristics associated with any type of data included volume, variety and velocity. These features formed a basis for the suitability of data to business applications. The volume of data handled under the big data concept is significantly larger than the conventional data volumes. Instead of the gigabytes of data conventionally handled in small data models, the big data concept has seen the realization of larger volumes such as petabytes to Brontobytes (Davenport and Dyche 2013). The implication of this has been wide application of data dependent operations such as those, which require cloud computing. The importance of big data in such contexts is thus undeniable. With high volumes of data handling, the concept of velocity also becomes known.
The velocity of big data describes the frequency of data handling which was initially intermittent. However, with the evolution of modern big data concept, data handling has adopted a more continuous approach, dependent only on the availability of data handling resources (Tufecki 2014). The present business environment requires dynamic data representations, which can only be achieved with big data. As opposed to the traditional set- up where data handling involved discreet procedures, big data offers the opportunity to carry out streaming operations, which result in a continuous flow of data. Moreover, due to the continuous mode of data handling and management that is required in social media platforms enables big data to be the mode of choice in social media businesses. The velocity of big data is even more pressing due to the continued expansion of data variety handled especially within the social media context.
Variety of big data relates the types of data that can be handled independently. Database management systems are restricted in terms of the data types they can handle. Similarly, small data contexts also handle significantly narrow range of data types while the big data concept enables handling a huge variety in the type and form of data. These key features form the basis of operation and applicability of all data concepts. Big data, offers these features and even more, which makes it the most preferred data model in businesses (Lycett 2013).
Besides, these conventional characteristics, other characteristics associated with big data include: exhaustiveness, high resolution, flexibility and relational characteristic. These characteristics enhance the benefits associated with data use, with the impact that a wide range of applications can be achieved through big data use. In terms of exhaustiveness, big data has the capacity to enable the recording of every single inactivity or activity that takes place based on data (Dalton and Thatcher 2015). This application has made big data essential particularly in the social media platforms, which require that all records be kept every second of every day. This requires immense storage space and large volumes of analyses. Moreover, exhaustiveness is also associated with accuracy and precision since every record kept has to be an accurate account of the events on the platforms. In addition, accuracy in timing is important for effective record keeping. Despite the advantage of big data in terms of exhaustiveness, the benefits of big data with reference to effective record keeping are still treated with concern. This is based on ethical considerations of the use of the data collected in record keeping. It is a question of ethical implications in the generation and distribution of data (Zikopoulos et al 2012).
Secondly, big data is associated with high resolution. In the contemporary data dependent operations, the variety of data that requires handling demands that high levels of resolution be available for effective representation and data management (Kitchin 2014). The range of data that requires handling includes reports, photos, visual and audio files. This variety in data requires that data handling procedures should result in the acquisition and management of data with high resolution. Big data is fine grained hence enables the display of data to be carried out with high precision and clarity. In the application of small data and databases, the representation of data depended on the use of unique identifiers, which enable data collation, sorting, monitoring, and creation of entity profiles. The unique identifiers were use to create relations between the extant data and the profiles of various entities who apply the data. This comes hand in hand with high data velocities, which are also associated with big data (Cukier and Mayer-Schoenberger 2011). The major challenge that data dependent businesses have had over the years has to do with the capacity to hold large volumes of data and to deal with large data transfers. Big data overcomes this challenge through an offer of flexibility and variety (Kitchin 2014).
In social media applications, the data handled ranges from posts, images, texts, videos and audio files, which requires high performance platforms. Traditional data models did not offer opportunities for the combination of various data varieties in single outputs. On the other hand, big data has overcome this challenge through the increase in management capacity, information extraction capabilities and data processing abilities. These capabilities have been associated with high performance big data platform infrastructure such as hadoops (Davenport and Dyche 2013). Hadoop uses various algorithms, which are conditioned to carry out complex instructions such as those related to text mining (Davenport and Dyche 2013). The importance of such a platform in social media is because big data makes it easy to create links between various data types. Big data links unstructured data types with the structured data types. This makes it the most suitable data model for application in social media platforms, a use which it has taken by stride (Lycett 2013). Moreover, the applications of big data in both the social media and in other platforms are also extensive as well as scalable. The flexibility associated with big data makes it possible to apply it various platforms, which experience operational changes and dynamism in data structures. In addition, big data can scale the walls of capability in terms of its provisions.
Human assemblages in Big Data application
While the major benefits associated with big data result from their non-human entities, the major challenges experienced in the application of big data arise from the human assemblages of big data. The socio-cultural concepts in the use of big data forms the greatest challenge as well as the greatest potential for the application of big data. These aspects include data generation and the mediation platforms. These assemblages contribute a lot in terms of ethical considerations in the use of big data (Tene and Polonetsky 2013).
Secondly, ethical issues also arise in the generation and application of big data. For instance, the mode of continuous data generation in social media raises concerns as to the nature of this data. The exhaustive nature of big data on the other hand, makes the scope of data generated wide. The implication of this is that regulations have to be laid down for restricting access to the generated data through processes such as encryption of data relating to personal information, or that requiring confidentiality (Beato et al n.d). In addition to this, the exposure of personal information by various social media platforms is also restricted and only carried out after authorization by the platform user (Hoadley et al 2009). A major concern that has been raised in the use of social media is the identification of information that could be considered private. This has led to the exposure of information that is to be held in confidentiality (Tene and Polonetsky 2013).
Furthermore, the political and economic conditions also influence the use of big data since data is considered as intellectual property. The use of such data is therefore subject to regulations guiding the protection of intellectual property and the relevance to intellectual property rights (Beato et al n.d). On the other hand, the applicability of big data in the monitoring of intellectual property information is also an important concept in the use of big data. This implies that the aggregation of big data into the intellectual property monitoring system can help in the creation of language commonalities, and thus drive communication improvement between specialists in intellectual property and owners of the IP rights (Swycher 2014).
Empirical Analysis: Facebook
Facebook as a social media platform has undergone immense growth since its inception more than a decade ago. The growth has been in terms of both organic reach and ideological expansion. Currently, Facebook has over 1.39 billion users worldwide. This implies that in every second, there are a huge number of users who are on Facebook. In the year 2014, it was reported that approximately 890 million users log into Facebook on a daily basis (Sedghi 2014). This figure has however risen in the past few months resulting in increase in the number of users per day. Moreover, these users also cover a wide range in terms of age and intellectual characteristics. This number of daily users can only translate into a higher need for data recording and handling capabilities.
The challenge in terms of capacity and capability is still inherent, needing the incessant application of big data. With more than 200 friends per user, Facebook continues to register immense activity in terms of interactions requiring recording, sharing of posts, messages, videos, and audio files. All these operations indicate the need for better data handling strategies that can only be addressed with big data. From the number of Facebook users per day, it is indicated that each of these users spends an average of 21 minutes each day on Facebook (Sedghi 2014). The implication of this is that the number of users of Facebook has grown exponentially, resulting in a significant growth of the social media platform’s net worth.
The organic growth in Facebook has been driven even further through the variation in the applications to which Facebook can be placed. For instance, from 2003, the increase in organizational use of Facebook for promotional purposes has been significant. The revenue obtained by Facebook from advertising has risen significantly as a result of the rise in Facebook application in advertising. From a value of $2.02 per user in 2009, the revenue has risen to an average of $8.05 per user in 2014 (Statista 2015). This means that with increasing use of Facebook in advertising, the variety of data that needs handling also increases. For instance, while posts and messages are mainly dependent on text data, advertisement may need the incorporation of other data types such as videos, audios and images. The implication is an increasing need for larger capacities and resolutions (Tufecki 2014).
Moreover, the use of Facebook has also expanded in nature of the devices used. From the fixed internet access that was traditionally confined to the use of Facebook. It has been reported that access to Facebook can be enhanced through access to the internet. It is approximately that at least 70 percent of internet users who are over 18 years of age have access to Facebook. This percentage increases with increase with age of the internet user. At 26 years of age, up to 84 percent of the internet users also access Facebook (Guimaraes 2014). The implication of this is that with an increase in age, the needs for socialization and interaction with others also rise hence the need to use big data.
Big data applications in social media
From the empirical analysis and the literature review, it has been established that the application of big data in social media are immense. This is based on the operations that have been described as being involved in social media. For instance, with increasing use of social media platforms such as Facebook, the need for better data management and analysis is required. First, in the use of Facebook, various data handling operations are carried out. The greatest role of the Facebook employees in using their information systems is to ensure that effective records are kept regarding the applications of Facebook. Over 9000 employees are involved in monitoring data transfer and managing other data related operations at Facebook. This means that the feature of big data with reference to exhaustiveness takes centre place in the effectiveness of the record keeping process. Big data, being exhaustive also helps Facebook to keep detailed records of all operations, which make it possible to produce statistics such as the revenue obtained from advertising per person annually.
These records can aid in monitoring organizational growth and charting the way forward for improved service delivery. Besides the general records kept to enable Facebook keep track of its operations, the company also has to provide usage records for individual users in the process referred to as datafication (van Dijck 2014). This implies that innovativeness and productivity in terms of technological advancements. This is because these records are so dynamic and cannot be followed up by individual employees.
The dynamic nature of the Facebook operations also requires that the company should take advantage of big data immensely. From the features of big data that have previously been described in the literature review, it has been established that big data is also associated with high velocities. The continuous nature of big data that enables streaming is very applicable in the context of social media platforms such as Facebook. This is because with the high number of users associated with these platforms on a daily basis, serial use is impossible. This implies that the data transfer and processing operations at Facebook have to be continuous daily and on a yearly basis. What this requires is the application of big data, which offers high velocity and enables a never-ending information flow. While this is critical and very beneficial to Facebook and to other social media platforms, it is also significantly costly in terms of management costs.
The diversification in application brings out a core concern in terms of data types. The data handled in advertisements is significantly of higher variety and requiring closer monitoring and management. At Facebook, the application of big data can aid in this concept greatly through improved resolution of the data transferred. Advertisement requires exceptional representation, which can only be achieved through high-resolution models. Small data and databases cannot be effective with respect to achieving high data resolution. Moreover, the fine graining associated with big data makes it possible for data to be collocated easily and to be sorted and monitored effectively. Sorting and monitoring data is essential in Face book’s record keeping activities, as the company has to produce records of revenue obtained from various operations. The sources of Facebook’s revenues include provision of communication platforms, sale of services and goods and advertisement (Kate 2014). In addition, collation of big data enables the association between textual information in adverts or in posts, which include images, and texts to be actualized. It is the role of the organization to ensure that data handling operations can be achieved without freezing.
One feature of big data that makes it possible for the organization to avoid freezing is that it is variable in terms of application. As has been confirmed, Facebook like other social media handles a wide range of data applications. The relationship between various data and media in the social network can be created or determined based on the application of big data. Data linking is often required in the social media platforms particularly due to variation of the data types. For example, through Facebook, it is possible to present various users as being related socially in real life. This requires that data relating to different individuals be liked together through the media. Moreover, it is also possible for individuals to create links to other media that is not within the social media platform in use. This has been applied widely in the advertising context where people provide links to product adverts instead of creating the adverts on social media. While this is beneficial to the Facebook users, it implies that greater attention has to be paid to the data that is in both sites. This requires exceptional management and processing capabilities, which can only be associated with big data.
An additional application that Facebook has for big data is in the extension of its operations. This is based on the flexibility associated with big data. This is based on the premise that with Facebook acquiring more social media platforms, the range of capabilities and operations also increases hence requiring a data format that can be extended to encompass new additions, and which can scale the heights of technological advancements.
The project has been effective in achieving the research aims. The first aim was to find out the features of big data that make it applicable to a wide range of sectors. It has been confirmed that besides being related to large data volumes, exhibiting high velocities and wide variety like other conventional data types, big data is also exhaustive, flexible, fine grained and relational. These features make big data essential in various applications, as it is possible to fine tune organizational needs to the various deliverables of big data. However, the main limitation experienced in the application of big data is associated with the high management costs. The benefits linked to big data however overcome this challenge in terms of incremental revenue. Secondly, big data has also been confirmed to be applicable in the social media in several aspects. The needs of social media operations vary in terms of velocity requirements, flexibility, and the need for high resolution. Big data offers all the solutions to dealing with the challenges associated with data use in social media. The applications of big data in the social media platforms therefore range from record keeping of user activities to keeping of actual records for specific users. Moreover, big data is also essential in the social media for creating relations, accessing information from other sites through information linkage and creation of flexible platforms that can be accessed as well as expanded at will. The growth of the social media use in the recent times has made the need for big data even more pressing as the variety of data applications also increase.
Finally, the roles played by big data in social media have been identified as including: enhancement of continuous flow of information, increase of platform flexibility, effective, accurate, and precise record keeping, enhanced data analysis, better data linkages between structured and unstructured data types, improvement of the handling of a wide data type range and better monitoring of data dependent operations.
From these findings, it can be concluded that the study has been beneficial and worthwhile in terms of knowledge provision. However, the main limitation that was faced during the study is the availability of limited information. It is therefore recommended that more research should be carried out in this area, particularly with reference to the benefits that particular social media platforms obtain from the use of big data. The need for additional information is very important and should be addressed by the academia as soon as possible.
Beato, F, Kohlweiss, M and Wouters, K n.d. Enforcing access controls in social network sites. Retrieved from https://www.cosic.esat.kuleuven.be/publications/article-1240.pdf
Cote, M 2014, Data motility: the materiality of big social data. Cultural Studies Review, 20, 1: 121-149
Cukier, K and Mayer-Schoenberger, V 2011, Is the rise of big data. Foreign Affairs.
Dalton, C and Thatcher, J 2015, What does a critical data studies look like, and why do we care? Seven points for a critical approach to ‘big data’. Society
Davenport, T and Dyche, J 2013, Big data in big companies. International Institute for Analytics.
Facebook 2015, Facebook newsroom. Retrieved from http://newsroom.fb.com/company-info/
Guimaraes, T 2014, 7 new statistics about Facebook users that reveal why it is such a powerful marketing platform. Business Insider. Retrieved from http://www.businessinsider.com/a-primer-on-facebook-user-statistics-2014-9
Hoadley, C, Xu, H, Lee, J, and Rosson, M 2009, ’Privacy as information access and illusory control: the case of the Facebook news feed privacy outcry’. Journal of Electronic Commerce Research and Applications. doi:10.1016/j.elerap.2009.05.001
Kate 2014, How do you make money from mobile? Retrieved from http://www.futureplatforms.com
Kitchin 2014, Big data, new epistemologies and paradigm shifts. Sage Journals. DOI: 10.1177/2053951714528481
Lycett, M 2013, Datafication: Making Sense of (Big) Data in a Complex World, European Journal of Information Systems, 22 (4), pp. 381‐386.
Madden, S 2012, From databases to big data. IEEE Internet Computing.
Sedghi, A 2014, 10 years of social networking in numbers. The Guardian. Retrieved from http://www.theguardian.com/news/datablog/2014/feb/04/facebook-in-numbers-statistics
Statista 2015, Facebook’s average advertising revenue per user from 2009 to 2014 (in US Dollars). Retrieved from http://www.statista.com/statistics/234056/facebooks-average-advertising-revenue-per-user/
Swycher, N 2014, ’Big data solutions to determining IP risk and value’. Intellectual Asset Management, 41-46.
Tene, O and Polonetsky, J 2013, Big Data for All: Privacy and User Control in the Age of Analytics, North Western Journal of Technology and Intellectual Property, 11, 5.
Tufekci, Z 2014, Big Questions for Social Media Big Data: Representativeness, Validity and Other Methodological Pitfalls. In ICWSM ’14: Proceedings of the 8th International AAAI Conference on Weblogs and Social Media, 2014.
Van Dijck, J 2014, Datafication, dataism and dataveillance: Big Data between scientific paradigm and ideology. Surveillance & Society 12(2): 197-208.
Vis, F 2013, A critical reflection on big data: considering APIs, researchers and tools as data makers. First Monday, 18, 10.
Zikopoulos, P, Eaton, C, deRoss, D, Deutsch, T, and Lapis, G 2012, Understanding big data: analytics for enterprise class hadoop and streaming data. McGraw Hill.