Developing a Recommendation Engine using Hybrid Approach
Client: A cloud-based interactive smart video platform that delivers personalized interactive videos on-demand to consumers. Objective Business: Increase time spent on site, number of videos watched, and repeat visits. The Engine: Devise an engine that offers relevant recommendation dynamically and in real-time. The Need Real-time recommendations need to be self-learning and based on massive amounts […]
MaxDiff modelling for product feature optimization
Objective To understand which product features or applications are most important to a particular market. Approach An online quantitative survey was conducted that included a MaxDiff exercise in which 40 different product features were evaluated on what is best and worst among 15 sets with 4 features in each set. MaxDiff modelling was used over […]
TURF to maximize messaging reach
Objective To understand the message bundle that makes up the most motivating brand story, differentiating from competition, and thereby capturing the largest portion of the market possible. Approach Message importance was derived from the MaxDiff design. The number of messages included in the design was 40. Message Bundle Evaluation was conducted by a physician/MD ‘carries […]
Market Mix Modelling/ROI Modelling
Objective Identify relationships between sales and various marketing activities to understand how various elements of the marketing mix combine to drive sales in a competitive framework. Approach From the client’s transnational database, we have considered the retail audit data like Volume sales, WARP, Weighted Distribution, and Promotion volumes. From media data, both GRP and NGRP […]
Understanding the impact of TV programmes on children’s skills
Objective Understanding impact of viewership on children’s skill levels (tested through an instrument devised by leading child psychologists) after controlling for location, socio-economic levels, gender, pre-exposure levels across a panel of over 5000 children across urban and rural India. Approach Using a naturalistic prospective longitudinal design, the study surveys children and their caregivers in several […]
Classification using higher order interactions: a hybrid approach
For representation in this blog, we have considered German credit data for classification of good and bad credit risk using the interaction of the input variables through a hybrid approach. In this approach, first we developed a decision tree for identifying the significant interactions of the input variables and then, a feed forward neural network […]
Power of Open data
Air-Pollution in Delhi reached a severe level during the festive season. Post Diwali the air quality in Delhi worsened and the city was covered in smog with zero visibility, reaching an alarming level with air pollution at its peak. While the city is smothered with air pollution there is an urgent need to […]
A silent move towards adaptive and model-dependent data models
Over the last decade there has been a substantial increase of the volume and complexity of the data we are collecting and storing. Now we see an increasing demand for real time data processing – a continuous process of input / output of data. Accordingly, the approach to data analysis and data processing has changed […]
Segmentation, Targeting and Positioning (STP)
Client: A consultancy firm in South East Asia Time: January 2014 ———————— Objective The objective was to develop a segmentation algorithm on the basis of a set of psychographic attributes like basic values, beliefs, attitudes, perceptions, motivations and needs (functional, emotional, social and situational) in order to embark on targeted marketing and to enable positioning […]
