-klZD--621x414@LiveMint.jpg "As much as 80% of existing data is also known to be unstructured, as it is primarily available in the form of speech, natural language and vision. Photo: iStock")
Cognition can be defined as the mental action or process of acquiring knowledge or comprehending thoughts, experiences, and the senses which are most critical but widely immeasurable: Sandeep Parikh, Partner, Intelligent Automation, EY
In the current digital era that we live in, advancement of technology has made it easier to collect, store and process huge quantum of data. But until quite recently, it was time consuming and difficult to make use of unstructured data for deriving time sensitive insights. The question was, can we somehow tap into this unstructured data on a real time basis and generate predictive insights that can assist in making better decisions?
As much as 80% of existing data is also known to be unstructured, as it is primarily available in the form of speech, natural language and vision. With digital technologies getting imbibed in our day to day lives, huge quantum of unstructured data is getting generated – from devices, social media, e-mails, videos etc. Unstructured data has no fixed format or structure and therefore it is hard to process it.
This data is the most critical source of insight (into individual preferences, into business performance etc.) that has the potential to enable an organization to take informed decisions. It is equipped to predict probable outcomes with greater precision. The data that is currently into play and used by most of the organizations is structured data, which is just 20% of the universal data pie and this is what corporates are currently tapping into.
This is where Cognition comes into the picture. In layman terms, cognition can be defined as the mental action or process of acquiring knowledge or comprehending thoughts, experiences, and the senses which are most critical but widely immeasurable. Systems and technologies that enable reasoning qualify as cognitive technologies. This also includes what is popularly known as Intelligent Automation – technologies such as RPA, Machine Learning, Natural Language Processing, Natural Language Generation and Artificial Intelligence.
The advent of Intelligent Automation started with the use of available data to automate mundane, rule-based activities. However, further advancements in Intelligent Automation led to an era of digital disruption wherein machines developed the capabilities to tap into unstructured data.
Cognitive makes it possible to understand complex content, engage in dialects in natural language, analyse unstructured information in conjunction with structured data, draw conclusions or take decisions based on structured or unstructured inputs such as reasoning, senses, verbal cues to arrive at the most probable outcome real-time. Key to deriving meaning from unstructured data using Cognitive is pattern discovery which uses Machine learning algorithms to analyse current data and predict future outcomes.
Examples of what Cognitive can do:
—Enable natural interactions with the humans using technologies like NLP, NLG, text analytics etc.
—Make unstructured data usable that can be processed and can be weaved with structured data
—Emulate humans to understand this data, make connections with the past data, form hypothesis, make inferences and come to a conclusion
—Enable continuous learning due to change in data set which may lead to different conclusion altogether
…and transforming businesses making them ready for the future
The capabilities of Cognitive have led to its increasing technology adoption across various industries and functions like customer service - where cognitive is assisting companies in constantly monitoring customer sentiments and alerting possible customer attrition. Companies today are widely using cognitive to unlock insights by analysing customer feedbacks and posts on social media, real-time, which is helping monitor and maintain brand image.
Apart from supporting and transforming business processes and the way business is done (like target selling, order management, procurement, payments etc.), some of the cases where cognitive can make / is making a real difference are as below:
Identifying criminals / criminal activities
Cognitive can assist security agencies by making connections between people, places, and their actions. They can identify if these analyses are leading towards any risk or criminal activity and send alert to take any preventive measures.
Looking ahead
Connectivity (based on Internet technology) is on the upswing. With this, quantum of data that is expected to be generated is going to increase at an unimaginable rate. Technology enterprises are investing in R&D to enable them to improve cognitive capabilities to churn out insights & decisions at faster rate and in almost real time. These enterprises are finding newer ways of processing various sources of data (speech, text, and visual data) and integrating them to generate better insights.
Businesses need to adopt Cognitive. They cannot keep away from this technology and still manage to stay relevant & competitive in the market. Businesses who have a better understanding of their markets, customers and looking to dynamically transform outcomes and experiences across in real time will survive in the long run and cognitive can assist businesses in this aspect.
