Widget Image
CURRENT ISSUE
ISSUE 00
SUPPORT US
PURCHASE PRINT
Follow us:
Saturday / April 10.
  • No products in the cart.

The Great Big Data Heist

[IMAGE: GETTY IMAGES]

It’s been a huge decade for Big Data and Artificial Intelligence, two of the biggest tech trends we have seen this century. From data-driven manufacturing to self-driving cars, we have witnessed dozens of jaw-dropping, previously unimaginable feats, all thanks to advances in Big Data analytics and AI. 

Not so long ago, businesses across industries often sat on tons of useful, game-changing data, unsure about the many ways they could put it to use to gain competitive advantage. But as methods in Machine Learning, Deep Learning, and natural language processing became more advanced while computing power went up, seemingly useless data suddenly began to make sense.

For instance, businesses could use customer data to analyse demographic profiles, shopping habits, and other behaviours, which helped improve marketing campaigns and overall customer experience.

Still, despite all the good that comes with AI, its growth presents a myriad of challenges for Big Data, especially when you consider how data-hungry AI systems can be.

These challenges represent the biggest roadblock that must be addressed before we can fully realize the potential of AI and Big Data.

 

Data Privacy and Security

AI systems, even the most basic forms, are usually very complex, with tons of algorithms obscuring what the system is actually doing under the hood. As such, any data used for such processing is usually hidden from view, which raises questions about transparency and privacyof such data.

Take, for instance, cookies, the pieces of code that are used to collect user data from websites for advanced analytics. While many countries now require websites to inform users about the use of cookies to collect data from browsers, there’s no way to know how much data or specific types of data that is collected via such websites.

Plus, there is always the issue of data security when AI systems are handling massive amounts of data across networked, distributed databases. In many automated industries such as the telecoms industry, stolen data, for instance, can be used to launch automated spam calls like robocalls, a popular nuisance in many countries globally.   

 

Limited Technical Capacity

Even though we have so far been successful at building faster and better processors for increased computing abilities, these abilities are constantly being challenged by increasingly demanding processing tasks and larger amounts of data to be processed.

AI algorithms are usually very complex, often requiring thousands of calculations sometimes even more computed every second. With the development of cloud and distributed processing over the past decade, it became possible to process such algorithms, ushering in the current age of AI-powered data analytics.

However, as demand for more powerful processors increases, bottlenecks will start emerging, making it difficult for enterprises to adopt the technology. For start-ups and small and medium businesses, this means raising huge sums of capital to bring on board better processors and larger storage servers, which many would struggle to do.

This trend also means businesses will have a hard time securing data across multiple, non-relational databases that are constantly evolving.

 

Lack of Human Capital

Data analytics is a complex field, a fact that gets even more complicated when you factor in machine learning, deep learning, and other components of AI that are often used to analyse data.

As such, there is a huge demand for data scientists who are talented in various fields, purely because the job is heavily multidisciplinary. One McKinsey study predicted that there would be close to 200,000 unfilled job positions across different industries for Big Data scientists and professionals by 2018 in the U.S alone. This demand will grow with increasing avenues for data collection and advanced AI-based analytical methods that would put pressure on organizations to find the right professionals to help handle such data.

In addition to Machine Learning and Data Mining, some of the skills that are required of Data Scientists include statistics, software engineering, linear algebra, programming languages such as Python and Java, and platforms such as Hadoop for advanced analytics.

 

Conclusion

As AI and Big Data continue disrupting industries across the board, issues related to transparency will inevitably force discussions into what people should really expect from AI-powered Data Analytics. For younger generations and digital natives who’ve often been said to be liberal with personal information on the internet, it is critical for organizations that use this data to clearly outline the scope within which such data will be used, lest legal issues arise from misuse of trust.