1 Case Study
Shrikaa
Canwetrustbigdata.docxRunning Head: CAN WE TRUST BIG DATA? 1
CAN WE TRUST BIG DATA? 6
Can we trust big data?
Abstract
Each business firm adopts IT infrastructure based on its business processes and organizational structure. This case study portrays the importance of bid data in business decision making processes and outlines its significance in business optimization.
Big data analytics has recently attracted the attention of both the academic and the business environment, due to the possibility of analyzing the underlying structure of big databases, in which commonly used statistical tools tend to fail (Elgendy & Elragal, 2016). In this regard, the managers of companies worldwide tend to consider big data analytics tools among the most promising tools in assisting the decision-making process at the company (Sivarajah et al., 2017). Through the better data organization, the big data analysis enables the manager to optimize the efficiency of the company by identifying the crucial variables that affect company’s performance and the market variability (Sivarajah et al., 2017).
Main challenges faced in big data analysis at Hadoop
Despite its broad acceptance in the past decades, big data analysis stills present multiple critical problems that may hamper its applicability in different real-life scenarios. The most common challenges are data challenges, process challenges, and management challenges. The analyst tries to overcome these challenges by the fair selection of the big data analysis tool (Gandomi & Heider, 2015).
According to Michael Walker , data challenges result from the intrinsic characteristics of data that face at Handop distributors and implementation of advanced technology. (Sivarajah et al., 2017). Thus, these challenges are related to the immense volume of available data that needs to be sorted out and analyzed, as Gantz and Reinsel estimated that the generated information by the end of 2020 would be as high as 40 trillion of gigabytes (Gantz & Reinsel, 2012). Other significant data challenges arise from the wide variety of data available, the speed at which they are generated, their accuracy level or the way in which they are displayed such that the analyst may understand any existing underlying trend.
Process challenges, on the other hand, group the different challenges faced by the big data analyst while processing the data (Sivarajah et al., 2017). These challenges include the multiple problems arising from the acquisition and safe storage of the gathered data, the grouping, categorization and integration of the different data, the analysis of the existing trends and the validation of the developed models.
Finally, the management challenges result from the need to maintain both the safety and privacy of the gathered data, the type of data that may be shared, the evaluation and recognition of the data ownership, and the lack of skills of most big data analysts (Sivarajah et al., 2017).
Big data analysis tools
The most commonly used big data analysis tools that will help in enhancing roductivity and increase salkesinclude text mining, audio analytics, and video analytics. The use of any of these tools is closely linked to the type of message containing the data. In this regard, the tools used for the extraction and identification of such data will vary depending on if the information represents a text, an audio or a video message (Gandomi & Heider, 2015).
In this regard, text mining focuses on the interpretation of the underlying information contained in text messages such as emails, blogs, forums or corporate communications (Gandomi & Heider, 2015). It quantifies the frequency of appearance of specific keywords, the extraction of information such as dates or names of people from a text, the automatic response to fundamental questions or the creation of a summary from the information contained in the document. Opinion mining is a particular text big data analysis tools that evaluate and categorizes the opinion that a person presents with a specific product or claim, thus being extremely useful in marketing research.
Audio analytics and speech analytics, on the other hand, focus on the interpretation of the information contained on an audio message such as a speech or a recorded phone call (Gandomi & Heider, 2015). Audio analysis tools are commonly based on the identification of specific combinations of morphemes, such as the -ing morpheme that indicates the word preceding it is a verb, thus representing a type of action. Similarly, video analysis tools focus on the identification of morphemes in a video message. However, in contrast with audio analysis tools, video tools profit from the presence of an image which may be used to interpret the existence of abnormal behavior (Gandomi & Heider, 2015).
Alternative specific big data analysis tools include the use of social, big data analysis and predictive big data analysis. The first is commonly used for the interpretation of data published in the social media platforms. For instance, the company may want to monitor the frequency at which its customers speak about their products on their Facebook profiles. On the other hand, predictive big data analysis tools focus on the application of statistic tools such as regression to a series of numeric data.
Conclusion
Big data analysis is instrumental considering how it helps the managers of the company gather, interpret, and analyze the information contained in an exponentially increasing amount of uncategorized data. For big data analysis to provide the necessary results, however, it is vital that the analyst selects the best analysis tool according to the type of data or information. In this regard, the methods vary significantly depending on if the information is contained in a text, a video or an audio message. Furthermore, specialized big data analysis tools are required to adequately interpret the data published in social media platforms or to analyze big datasets of numeric data.
From this point of view, big data analysis, although useful, is a very complex process and presents multiple challenges for the analyst. In this regard, the main challenges faced by the analyst result from the immense amount of available data, their variability, the speed at which they are generated, and the requirements to (i) select the appropriate analysis tool, and (ii) ensure the safety of the data to prevent data breaches.
References
Gandomi, A., & Haider, M. (2015). Beyond the hype: Big data concepts, methods, and analytics. International Journal of Information Management, 35(2), 137-144. doi:10.1016/j.ijinfomgt.2014.10.007
Gantz, J., & Reinsel, D. (2012, December). The digital Universe in 2020: Big Data, Bigger Digital Shadow s, and Biggest Grow th in the Far East. Retrieved February 26, 2018, from https://www.emc.com/collateral/analyst-reports/idc-the-digital-universe-in-2020.pdf
Sivarajah, U., Kamal, M. M., Irani, Z., & Weerakkody, V. (2017). Critical analysis of Big Data challenges and analytical methods. Journal of Business Research, 70, 263-286. doi:10.1016/j.jbusres.2016.08.001
