Experts caution banks, telcos for not analyzing unstructured data
The adoption of digital banking by banks and fintechs as well as digital transformation by other service oriented organisations as a key strategy have raised the need for them to extend analysing of unstructured data, experts have cautioned.
They said: “This means opening up of their platforms to third party vendors/Partners, using multiple channels to offer services to their customers like social media, Mobile, Web, PoS etc. These generate massive volumes of data and expose the organizations to threats Presently, there are two major kinds of Data known as Structured and Unstructured Data, and a third one that sits between both data types called Semi-Unstructured Data.
Ethelbert Mbama, chief executive officer, Best of Breeds Business Solutions Limited, said unstructured data have its own internal structure, but does not conform neatly into a spreadsheet or database. Most business interactions are unstructured in nature.
“Today more than 80 percent of the data generated is unstructured and it keeps increasing due to the digital transformation agenda of most companies and opening up of channels and platforms. The fundamental challenge of unstructured data sources is that they are difficult for nontechnical business users and data analysts alike to unbox, understand, and prepare for analytic use. Thus, these data are mostly ignored and not used in analysis of events.
“This will not give the Business, a true picture of events. Examples could be sensor data (huge in percentage), social media streams, Log files, images/pictures, videos files, audio files, mobile data, Text data, Chats messages, PDF files, Word Documents, Emails, Webpages, etc.
“This represents about 80% – 85% of data that organizations generate today. In fact enterprises are investing heavily into data analytics of unstructured data as its major chunk is getting unnoticed.
“Outside the Handling of Fraud and Cyber Threats, Organizations that has the platform that can handle all kinds of data, analyze them, can be a step ahead of their competitors in all its operational engagement because they have 360 degrees view of the entire operations, Behavioral patterns, profiling and can, thus, use Real Time predictive analytics to make accurate predictions and decisions.
“Dealing with only Structured Data limits you to analyzing, only 20% of your entire Data. The remaining 80% is where most information resides as you adopt digital engagement, multi channels and Data explosion,” he said.
Omar Yaacoubi, chief executive officer and co-founder, Barac- fraud and cybersecurity solutions provider, urged banks and telcos to step up analysis of unstructured data which has become a source of fraud and cyber-attacks through innovative solutions that can adapt to hackers behaviour.
“Sometimes they change their behaviour so the solutions the banks use needs to change their behaviour as well. With the current solutions that they have especially with the rule base solutions, banks will not be able to do that because it takes up to six to nine months to change the rules where you have to modify rules, it is very complex this is where solutions with behavioural analytic and machine learning will help to solve the problem.
“More so, opening those silos that the banks have where the transactions system works alone, CRM system works alone, core banking solution will work alone, all those system are unique and will work alone on their own, so that is where they need solution that communicate, they need solution that have holistic view, they solution that can detect attacks across the whole channels that they have especially as they move into cloud and adding new capabilities and opening their networks to new features like USSD, Fintech companies that have access to the banking data.