As the use of big data increases, challenges such as invasion of privacy, data ownership issues, and lack of objectivity in analytics arise. It’s worth discussing whether we can maximize the benefits of big data analytics without addressing these issues.
People generate huge amounts of data every day without realizing it. Facebook, one of the social networking services (SNS), accumulates tens of billions of posts every month, and messenger apps like WhatsApp, Zalo, and Telegram send tens of billions of messages a day. In addition to these posts and messages, search queries, purchases, web page visits, blog posts, and more all accumulate as data. According to the MIT Business School, we generate around 250 bytes of information per day through mobile devices, social media, online commerce, and more. We’re generating more data than we can handle, and it’s called big data, a term that’s evolving to encompass more than just big data, but the process of analyzing and leveraging it.
Big data is not a new concept; there are already many successful examples of companies that have utilized big data, including Google, Netflix, and Apple. Google has been able to forecast the flu faster than the CDC by analyzing the frequency of searches for fever, cough, and more. It has also improved the accuracy of its automatic translator by statistically comparing billions of documents. Netflix developed Cinematch, a service that analyzes members’ tastes and recommends movies based on what they’ve watched. Apple’s voice command software, Siri, is another example of big data in action. When a user asks a question through Siri, the data is sent to Apple’s main servers, where an artificial intelligence algorithm at Apple’s headquarters analyzes the question and sends the answer to the user. The AI algorithm is built on a huge amount of data, and as more questions are sent to Apple’s servers, the database grows stronger and Siri’s responses become more sophisticated. In this way, big data can uncover new information, which, depending on how it’s processed, can create an infinite amount of value. But is it just a matter of applying big data analytics here and there? There are still significant challenges to big data analytics that need to be addressed.
First, a major challenge with big data analytics is the issue of personal privacy. The ultimate goal of big data is to cover everything, which means looking at everything that happens everywhere as data. In fact, widening the scope of data by dataizing all the things we used to pass by can bring about great change and progress. For example, imagine a service that collects data by measuring a person’s biological information and behavioral patterns in real time, and then uses that data to sound an alarm before a person develops high blood pressure or a stroke. The data that could be considered could be anything from what a person eats, real-time blood pressure, bathroom visits, gait, sleep patterns, and so on. If a simple chip could be inserted into the body to collect this data in real time and send it to an analytics center to predict illness, would this service be viable? While the idea of being able to predict illness is certainly appealing, most people would be skeptical about allowing all of their behavior and body information to be analyzed. In the previous example, individuals can choose whether or not to provide their information, but many people already do so without realizing it. People post on social media simply to communicate, but those posts are analyzed and used to understand people’s needs, and to find out how they react to and improve products. However, just like the protection of publicity rights, which requires that photos taken without a person’s consent not be used, just because analysts have access to social media posts doesn’t mean it’s okay to collect and analyze them. There are privacy concerns, and legal clarifications or agreements between individuals, sites, and analytics companies are needed.
In addition to privacy concerns, there is also the issue of unclear data ownership. People don’t know how long companies keep their data, to what extent companies can re-process it, and whether individuals have the right to completely delete their data once it’s in their possession. If a person’s data is moved to a U.S.-based data center through a cloud service like Google or Apple, who owns that data? The debate between individuals and companies over the ownership and use of data is unlikely to end anytime soon. Moreover, data is crossing borders, so issues such as who owns data and how much information should be made public need to be discussed on a global level, not just within a country.
Finally, big data analytics is inevitably not completely objective. In the past, there were assumptions that had to be made due to a lack of data, but now that we’re dealing with so much data, we don’t need to make those assumptions. This makes data analysis less subjective than it used to be, but it doesn’t mean that big data analytics is highly quantitative and objective. Depending on the topic or purpose of the analysis, the analyst’s subjectivity will come into play, starting with what data to work with. Even if you collect all the data you want, the initial data you get will inherently contain outliers and unnecessary values. The process of determining this and refining the data for the actual analysis will inevitably involve subjectivity, especially when it comes to identifying the most important implications for the analysis. This subjectivity can compromise the true meaning of the original data, which can defeat the purpose of big data analytics in the first place: to help analysts extract valuable information from the original data to better classify and even predict the data, not just to get results that fit the analyst’s subjectivity. Therefore, it is inappropriate to assume that big data analytics is quantitative and objective just because it is based on a lot of data. You should recognize the subjectivity of your analysis and look for ways to increase objectivity.
Big data analytics is emerging as a powerful tool in a variety of fields, and successful examples point to a bright future for big data analytics. However, big data analytics brings with it the aforementioned issues of individual privacy, data ownership and licensing, and objectivity of analytics. Without a concerted effort to address these issues, big data analytics will continue to be limited.
