SAS AND NMIMS LEVERAGE THE POWER OF DATA TO SOLVE REAL WORLD CHALLENGES
Collaborate to organise India’s largest analytics championship to work for the good of humanity
The need of the hour in the world of big data is to understand how to make the best use of it. With enough data, patterns begin to form which can detect anomalies. Complex and abstract problems can be broken down and solved using the available data. With this principle in mind, SAS in collaboration with NMIMS organized the Business Analytics and Data Mining (BADM) championship sponsored by HUL and CEAT to solve real world problems. The participating students had to work with the data given to them and help solve a plaguing problem in society - the increasing rate of student suicides due to depression. The challenge given was to determine the characteristics of students that are most likely to suffer from depression and detect patterns and suggest prevention measures for stakeholders with analysis using SAS tools. This year’s championship also saw participation from working professionals who were given data of HUL skin lightening products to conduct a data analysis for sales projections. The championship marked the second year of collaboration between NMIMS and SAS for this competition. It is India’s largest Analytics Championship which is moving in a direction to become Asia’s Largest Analytics Championship. Currently SAS is organizing similar championships in China & Taiwan in addition to India. Winners of each country have been endowed with the opportunity to participate and learn more on analytics from global experts at the Global Forum at SAS Headquarters in Cary (US). This year saw a 3-fold increase in the participation of students as compared to the year 2017, bringing in 2963 analytics enthusiasts pan India from colleges like IITs, IIMs, NITs and many esteemed institutes; an approximate of about 525 colleges across 25 states in India. At the end of two rounds of filtration, a total of 217 students went ahead to become a part of the 24-hour hackathon held at the NMIMS Mumbai campus. The championship began with a keynote address given by Dr. Rajan Saxena, Vice Chancellor, SVKM's NMIMS Deemed to be University. He spoke about how it is crucial for people to learn and to use data as a key asset for decision making. He also stated how management schools and universities such as NMIMS are focusing on this aspect by equipping their students with skills like analytics, decision making and diversity management. Dr. Catherine Truxillo, Director of Advanced Analytics Education, SAS came forth to share insights on how analytics is shaping the future and empowering the people to take better decisions. She stated, “All teams are gathered here together for a common goal to use data for greater good. Today we will work on a very pressing yet highly relevant issue on a mental health condition prevalent amongst many students. More than one in ten students in India suffer from clinical depression. By working on this competition, we will be helping to build relief and prevention for this social issue. We are all involved here coming together to be a part of the Data for Good movement, which encourages using data in meaningful ways to solve humanitarian issues.” After a series of enlightening keynote addresses commenced a panel discussion with stalwarts from the industry discussing the usage and implications of data science and analytics within companies to address issues faced in real time. The panel consisted of Abhijit Varma, Senior Partner - KPMG, Paresh Goel, Head of Decision Science - IndusInd Bank, Ritesh Srivastava, Advanced Analytics leader - Novartis, Susan Mani, Data Scientist - Analytic Edge Pvt. Ltd., Manish Desai, Sr. Regional Director Consulting - SAS Asia Pacific, Mritunjay Bhanot, Head Business Analytics Retail Lending and Payments - Axis Bank, Bireshwar Chatterjee, VP and Head Western Region - NSE and was moderated by Mr. Amul Desai, Director - NMIMS’s Centre of Excellence in Analytics/Data Sciences. The panel spoke at length about how data science and its applications have been steadily changing the way business is done and the way we live. They also addressed that considering 90% of all of the world’s data has been created in the past few years, there’s a lot of growth ahead in this exciting field. After the panel discussion it was time for the participants to take their positions in the lab where they spent the next 24 hours solving the problem statement using SAS tools to come up with solutions and insights.
Insights from the world of data
Analytics isn’t just a tool or technology; it is a way of thinking. That was the driving idea for this 24 hour hackathon - how participants approached problems in a limited time span. At the end of the 24 hours, the participants submitted their analysis reports to the jury and 5 groups from the student category were shortlisted to present their findings. The students were given an interesting problem statement on a very current humanitarian issue pertaining to depression levels and suicides. They were allocated 2 datasets: one on the number of suicides for various ages, profession groups for men and women across the states over a time span of 14 years. And the second on depression levels in students given their demographic and behavioral patterns collected through National Crime Records Bureau (NCRB) survey of around 14,000 students. All this data was spread across many files with different formats and irregularities. This was broken down categorically to identify which were the most significant characteristics of students that drive them towards suicidal tendencies and to find out the age group of the students who are more prone to depression and suicide. They addressed various questions like which demographics were more prone to suicide, which states had the maximum case of suicide reports and what major factor in students affected the depression level the most. The results pointed out to how the rate of suicide among the students is highest in Maharashtra followed by Tamil Nadu and West Bengal. The data analyzed in the competition concluded that factors like smoking, peer pressure and physical exercise are highly correlated to depression. Males who smoked were more prone to depression than females. Standard scales like Hospital Anxiety and Depression Scale (HADS) were used to determine which category of students was more prone to take serious steps like committing suicide. On using the predictive modeling approach which uses data mining and probability to forecast outcomes, it was noted that the mental health of students highly depended on their exercise routine, physical activity, IQ level, peer pressure and education level. These findings helped in outlining some possible measures and preventions for the stakeholders. It was suggested that as much as universities lay emphasis on education, they should also emphasize on increasing physical activities as they aid the release of endorphins which help in elevating moods; cultivating an environment that is supportive and helps students become vocal and increasing social engagements apart from cultural activities as an important part of the curriculum. An environment of cooperation and collaboration rather than competition shall lead to better mental health and less cases of student depression. Although all the teams managed to do an in-depth analysis, there were three top teams (IIM Calcutta, MPSTME (NMIMS), NIT Hamirpur) that stood out with their well thought models and approach to problem solving with the first place winners taking home a cash prize of INR 2.5 Lakhs and a sponsored trip to the USA to be a part of the SAS Global Forum.
Solving real life business problems
Apart from the student championship, this year the hackathon was also open to corporates across India. The participation was overwhelming with a total participation of 237 working professionals from companies like Idea, TCS, Core Compete, Mahindra & Mahindra, Bajaj Allianz, Franklin Templeton, and many more. After an initial shortlisting round, 88 participants went ahead to be a part of the 24 hour hackathon. Mr. Manish Makhijani, Head Consumer & Market Insights - HUL addressed the participants about the use of data analytics in the world of business and how it is a reality that we need to incorporate into our lives. He spoke about how analytics has changed the way digital world works. He stated, “By 2019 it is predicted that computers will have emotional intelligence which is same as humans. By the middle of the century, computers will be a billion times more powerful than human brain. If you understand analytics, if you understand data if you understand what it makes sense of, you'll have a job in the future.” The problem statement given to the corporates was from a business perspective. They were tasked with data from HUL for skin lightening products in the mass (Fair & Lovely) & premium segments (Ponds, Lakme). The objective was for the participants to understand the consumer behavior for these categories of products across the states and to project their market opportunity, price point, market share and overall ability to capture the market till 2020. The final 5 teams shortlisted for presentation spoke about how the marketing efforts needed to be customized for each segment based on what worked and what did not, in order to increase the ROI of campaigns. In the highly competitive market of mass skin lightening products in rural segments where price margins were already cut throat; analysis indicated that brand equity became the deciding factor for sales volume. There was a high correlation of sales with the employment levels, proportion of female population and presence of banking facilities in a given area. After all the insights and contemplation by the jury, top three winners declared were from RBS India, Bajaj Allianz Pune and Accenture. The Championship also attracted participation from large organizations such as LTI, HDFC Bank, Axis Bank, SAS, ETCS, Kotak Mahindra Bank, Tech Mahindra, LogisticNow, RaceNext, Xorient and CRG Solutions in the job fair organized alongside the championship offering lucrative internships and job opportunities for students. What was more important was that the championship was able to illustrate the fact that how often we think about big data as a purely analytical effort, but sometimes it can be more useful in improving communication and transparency. By understanding the implications and real world uses of data science, we can help solve persistent problems plaguing our daily lives.