The abilities of artificial intelligence today are not hidden from anyone. It has transformed companies across all industries, as businesses are pouring huge amount of capital on it. The technology has always been in its experimenting phase indicating a robust growth in AI products and applications in years to come.
Artificial intelligence is not limited to just IT or tech industry, it is being largely utilized in other areas such as retail, healthcare, business, education, law, and manufacturing. As advancements in AI technologies are on its swing, it is becoming more interesting how organizations are implementing this tech for effective business and customer experience.
Broadcast Audience Research Council India (BARC India) is one such company that is proving itself quite pragmatic by leveraging AI and machine learning to gain productive outcomes. In an exclusive interview with BARC India EVP – Product Leadership & Technology, Vivekanand Jha explains how artificial intelligence has been a tremendous breakthrough for BARC and for its customers.
AI in healthcare, retail, construction, cyber security etc, you name it and AI is present in almost all the industries/ sectors possible. How would you define the use of AI by BARC India?
From a T.V. ratings provider to a company of insights helping businesses, it has been a path-breaking journey for BARC India. We, as a company are constantly evolving – from launching unique products to improving client servicing to implementing cutting-edge technology. At the heart of everything BARC pushes for automation through machine learning and Artificial Intelligence (AI). This is a necessary component of our work due to the sheer volume of our data. We have TV viewing individual and House Hold (HH) level data with Panel Demography. AI is used for both capturing the data and to process the data more efficiently by applying human intelligence on large volume of data. We are using AI for speech recognition/Image recognition in different applications and products. AI helps in detecting heterogeneous genre and content tastes (over TG and over time band).
How big a role does AI play in your business? In other words, in your business, where accuracy is everything and companies depend on your insights to set their ad-spends, how big would you say is the role of technology in making these predictions?
Technology is key. The data which BARC India collects, compiles, and integrates provides examples of the 5Vs – volume, velocity, variety, veracity, and value. Leveraging such data is not always a straight forward decision. Collecting data from 180,000 individuals’ daily and integrating massive proprietary databases and alternative data sources requires a means for the data to be stored in an efficient, accessible, yet cost effective manner. At the same time, with data this large and variable, human intervention needs to be minimized, therefore requiring usage of the latest in AI and machine learning technologies to automate processes in an accurate fashion.
We at BARC, rather than serving as a replacement for human intelligence and ingenuity, take Artificial Intelligence as a supporting tool. We are adept at processing and analyzing troves of data far more quickly and getting synthesized courses of action. In this way, we are using AI to help required output possible consequences of each action and streamline the decision-making process.
Our data serves as the currency for the Indian broadcast television industry. It is through this data that broadcasters and agencies make key data-driven decisions relating to programming, strategy, targeting and planning. Providing the data in usable forms in a timely fashion is highly dependent on technology. Improvement is a constant journey at BARC while we continue exploring different technology
In your business, where Data is King and everything depends on data, how do you ensure data is safe and secure and not manipulated or tampered?
We have implemented enterprise level security architecture with robust perimeter and access controls measures and the entire data path from source to delivery is end to end encrypted. Additionally we are testing with MDM (mobile device management) on all devices used by field force for data collection and survey to ensure end point security controls are in place. Provides single sign-in for all corporate applications (on-premises, SaaS, IaaS) and PIM (privilege identity management in place) and also relevant access controls in place. We have multi-layer firewall and our data fully resideswithin India. Data stored are in code based and its master mapping are stored in system and refereed when required. IT systems and general controls as well as security are validated by renowned security experts from time to time.We have kept on learning and improving on these aspects over a period of time.
As a measurement company, what is the common difference you see between AI and ML?
AI and ML are the two clichés yet trending and buzzing topics these days. The fact is that many people tend to understand and consider AI and Machine learning as the very same thing. Both the topics are elated to Big Data and Data Science. ML is the subset of AI which addresses the simulation of the intelligent behavior in computers. It helps us to imitate intelligent process defined by business.
How much of an influence does AI actually have on the data crunched by BARC India?
As mentioned earlier, our data serves as the currency for the Indian broadcast television industry. It is through this data that broadcasters and agencies make key data-driven decisions relating to programming, strategy, targeting and planning.
Data Crunching for BARC India is about getting valuable insights from the data and sharing it with the industry to add value to their businesses. Turning hidden data into actionable information with ability to influence businesses and process decisions is our practice as a data driven organization. AI combines big data and ML functionality enables IT operations to deliver robust and accurate data to market. For example – Taking a trend of multiple week, we are identifying the exception basis rule defined by users. We then create a tool to hash the variance and threshold and then using this information we can apply ML. The tuning of rules creates the actionable on dashboard and hence help used to deliver data
Are AI applications being used to foresee and predict trends in the broadcast sector?
Numerous media firms these days are adopting a data first approach to aid growth. What are your thoughts on this?
Data is the new OIL. Decision making on gutt used to play a large role in media ecosystem earlier, and while it will continue to do so, it will be aided with more and more data. With data being so pervasive and all available these days, the key is to identify the right data set and integrate that with business decision making leading to a successful growth of the organization
What are some of the challenges overcome with the adoption of AI and ML?
Faster Analysis Prediction and Processing, reduces manual/ human dependence, improves precision of business rules and models, easy outlier and anomalies detection and recommending the right product basis analyzing product utilization trends are some of the challenges that can be overcome with the adoption of AI and ML.