5 Myths about Data Analytics and Artificial Intelligence to Think About



Big data and AI are buzz words. There are increasing interesting discussions how big data and AI have changed human lives. But there are also myths that have prevented companies from adopting data analytics and AI. Does your company use data analytics and AI?


“You can have data without information, but you cannot have information without data.” - Daniel Keys Moran

Few dispute that companies have more data than ever at their disposal to discover hidden insights. Using and understanding data is a crucial competitive advantage.


 

Myth 1: Data analytics is only for large companies


Data analytics is not only for large companies. With 1,000 lines of data, you can carry out meaningful analysis. Any type of company needs data analytics to gain a competitive advantage, as the overall business environment becomes more competitive .


For example, smaller companies can analyse social media conversations and buying trends. It helps identify customer needs and quickly capitalize on emerging sales opportunities.


It is more critical to ask good data questions and understand data quality. Not all data is created (or stored) equally. And vast resources does not guarantee success, but quality resources does.


 

Myth 2: Data analytics is extremely costly


Data analytics is becoming increasingly affordable, especially with cloud-based solutions. Companies do not need to set-up infrastructure without fully utilising it or to use data scientist for every task. Cloud based solutions also provides flexibility where it allow users to pay only for the capacity that they use.


The cloud-based solutions can be used for nearly all systems and processes. This has tremendously brought down cost for computing power and data storage. Processes which require less predictive accuracy can use off-the-shelf models. There are also many free open-source tools available to analyse and visualise big data such as Python and R.


Sometimes, there is no need to hire specialized expertise, such as a Ph.D.-level statistician or data scientist, before it’s necessary. It can also contribute to higher staff turnaround. It is often easier to hire these specialised talents than retaining it, since the statistician or data scientist who longs to make a positive impact on the business but can’t, will leave the company.


 

Myth 3: Data analytics (and AI) can solve all problems


Data analytics (in particular AI) has opened up a wealth of promising opportunities. But the belief that given enough good quality data, AI can solve all problems is not true. AI still does not have a full reasoning power like a human, that is it cannot think as to why a particular method is happening that way or ‘introspect’ their own outcomes.


AI needs to be able to explain itself, to represent to its users what it knows, how it knows it, and what it is doing about it. Then it can have a symbiotic relationship with humans. AI also has contextual limitations. It does not have sufficient perception of the user’s environment, situation, and context to reason properly. In future, this may change, but with its limitations, it still cannot solve all problems without human. This links to Point 5.


 

Myth 4: Data analytics can always accurately predict the future


Another myth often heard is predicting the future is easy with analytics. This isn't true. To predict the future, a data management strategy must be in place, otherwise predictions may give false confidence. Predicting the future uses probability analysis and based its recommendations on it.


For example, Netflix launched a $1 million competition to improve its recommendation system by 10%. It paid the bounty for the solution which came about 2 years later but later ignored the code. The enhanced algorithms “did not seem to justify the engineering effort needed to bring them into a production environment". The algorithm was built in an environment which has changed, such as moving from DVD to online streaming.


 

Myth 5: AI can take over all human jobs


AI works well with helping human with their jobs, but it is unlikely AI can take over all human jobs. The machine can learn from data but it still needs human to validate assumptions and rules made.


But when the rules are verified, AI has the potential to make more accurate predictions than human. For example, AI is reported to be able to detect breast cancer more accurately than humans. AI also dramatically improves healthcare as it increases access and takes over more routine tasks from human.


But in the end, doctors still need to look at the diagnosis as machines do not and cannot verify the accuracy of the underlying data they are given to learn from. It still takes human to design and train the AI model.


 

We live in a world now where data is generated from everything - smartphones, cars, wearables, appliances, social networking sites. Data can provide significant value. In this article, we addressed the myths on data analytics and AI, so that more companies will benefit from its adoption.


You can start by understanding areas of your business that can benefit. It is not necessary to go for a full adoption when starting, but adoption can take small steps to reach the end goal of seamlessly using advanced analytics and artificial intelligence in everyday activities of the business.


Do you agree that any company can adopt data analytics and AI? Share by leaving us a comment. If you require more information or assistance on digitalisation activities, contact us. We want to be an extension of our clients. Subscribe to our newsletter for regular feeds.


Did you find this blog post helpful? Share the post! Have feedback or other ideas? We'd love to hear from you.


 

References


Inside Big Data, Accessible and Affordable Big Data for Small Business, https://insidebigdata.com/2019/10/13/accessible-and-affordable-big-data-for-small-business/, published 13 October 2019

The Business Journals, Big Data Analytics No Longer for Big Companies, https://www.bizjournals.com/bizjournals/how-to/technology/2015/11/big-data-analytics-no-longer-for-big-companies.html, published 20 November 2015

76 views0 comments