Throughout the past decade, technology journalists have been under fire from readers with accusations of promoting fake news. You must forgive us if we have become just a little bit defensive. As Brien Posey points out in his article from March 2018, for the most part, this is a “tin-foil hat conspiracy.” In many political situations, “fake news” has become a meaningless and overused buzzword used to denigrate real news that may be embarrassing or distressing. In technology, many use the term fake news to criticize overly sensational articles published mainly to grab attention and revenue — yes, click baits.
Nonetheless, we cannot discount the fact that fake news exists in journalism, and technical topics are not immune. The difference is that we have the power to do something about it in the world of technology. Armed with this knowledge, along with the substantial intel gathered from the recent political era of fake news, data engineers have been busy analyzing the data collected and stored with the goal of building algorithms that can reveal patterns and trends associated with fake news. The lessons learned from this process have been of great benefit to the world of Big Data. Here are five truths in a world of fake news.
Not all data is equal
For a couple of decades, we have been busy collecting and storing massive quantities of data. However, it is only recently that we, at the enterprise level, have captured the ability to effectively leverage that data for the betterment of the enterprise. Big Data is about extracting and analyzing large volumes of data to reveal patterns and trends. This journey has not been without controversy. End-users have been frustrated by being tracked and targeted by savvy marketing campaigns. This has impacted the quality of the data as users intentionally misrepresent input in an effort to maintain some element of privacy. Therefore, we have to make an investment to build algorithms that can compensate for this known issue. Analyzing rough, unfiltered, and potentially fake data may not supply accurate information to make major business decisions.
Fake news loves emotional responses
Every stimulus we experience as human beings will result in us feeling and expressing some type of emotion. In part, this is why social media has experienced such massive success. We love to express our emotions, and we want everyone to know. Fake news thrives on this one human trait that we all share. Fake news alerts are intentionally written to invoke an emotional response in the hopes that we will share what we think is our own opinion. In reality, we have been manipulated by cleverly written words. And it works. We often react without taking a moment to realize that there may be more, or less, to the story. The result is that we feel compelled to express our emotions, and we want everyone to know. Because this is a repeatable behavior, Big Data analysts have been able to integrate code within their algorithms that can detect and filter data that instigates an emotional response, thereby filtering out what could potentially be fake news. This leads us to No. 3.
Optimization of information sharing
Fake news is not unlike The Borg from Star Trek. It needs to grow, and so we must be assimilated. This is how fake news is shared and can move rapidly through the ether world. We basically live with two kinds of data. The first is data that we need to retain. This is important data that can help us to make important personal and business decisions. The second type of data is that which we want to discard. Of course, both categories can be further broken down, but you get the idea. What we don’t want to happen is to use the second type of data to make important decisions. Unfortunately, this is often the case as we absorb massive quantities of fake news daily, often without even realizing that it is fake news. In the business world, Big Data has grown to have the ability to filter these two categories based on specific criteria. This helps to ensure business decisions are made using data that is relevant and proven.
Be very aware of bias
Bias is defined as an unfair and preconceived opinion about someone or something. Bias can influence a business decision when the data in question is not representative of the area of study. We all have biases as a result of our perspective, which is usually based on our personal lived experiences. Bias can unfavorably influence decisions, and the writers of fake news have learned to use bias to nudge pop culture in a specific direction. Filters used in the science of Big Data are learning to integrate the ability to filter based on levels of bias within the data gathered.
The role of data scientist is a great career path
We play a lot of games with job titles and classifications. Data scientist, data engineer, and data analyst are all job titles that can be associated with the career path for the dissemination of Big Data. For ease of reference, we will use the title data scientist. Historically, organizations have been frustrated by the process of storing large quantities of data with no ability to extract valuable information. Even worse, unfiltered data is used to make business decisions. When this happens, there is no way of truly understanding if there is value in the conclusions that are made. But there’s a new sheriff in town. Enter the latest hot tech career path. For the technically minded and the detail-oriented, fake news has supplied the world of IT with scads of data to use in testing the accuracy and viability of our algorithms. The result is a very lucrative and interesting career path.
While we continue to learn as end-users to keep our personal information close, the enterprise has learned how to leverage lessons learned from fake news to debunk and disseminate data. The ultimate goal is to be left with only the data that we want to keep and use to make those important business decisions. This is important in the world of business. Fake news moves fast and can quickly erode the trust we have built with our customers, shareholders, and employees.
Featured image: Shutterstock