Analytics Magazine Analytics Magazine, January/February 2014 | Page 29

most cost-effective opportunities for increasing their sustainability. IBM: “A SMS text classification system for UNICEF Uganda” Authors: Prem Melville, Vijil Chenthamarakshan, Rick Lawrence, Solomon Assefa, Machine Learning Group, IBM Research, Yorktown Heights, N.Y.; James Powell, Sharad Sapra, UNICEF Uganda; Rajesh Anandan, US Fund for UNICEF, New York, N.Y. Presenting author: Rick Lawrence Abstract: U-report is an opensource SMS platform operated by UNICEF Uganda, designed to give community members a voice on issues that impact them. Data received by the system are either SMS responses to a poll conducted by UNICEF or unsolicited reports of problems occurring anywhere within Uganda. There are currently more than 200,000 Ureport participants, and they send up to 10,000 unsolicited text messages a week. The objective of the program in Uganda is to understand the data in real time and have issues addressed by the appropriate department in UNICEF in a timely manner. Given the high volume and velocity of the data streams, manual inspection of all messages is no longer sustainable. This talk describes an automated A NA L Y T I C S message understanding and routing system deployed my IBM at UNICEF. We employ recent advances in data mining to get the most out of labeled training data, while incorporating domain knowledge from experts. We discuss the tradeoffs, design choices and challenges in applying such techniques in a real-world deployment. We conclude with a discussion of the societal impact that U-report is already driving in Uganda and discuss plans for future deployment. Fiserv: “Price & revenue optimization for one of the largest acquiring banks in South America” Authors: Suman Kumar Singh, Aditya Khandekar, Tarang Goyal, Fiserv Abstract: The Merchant Acquirer had about a million merchants in mass market. An opportunity existed to selectively optimize price for merchants based on certain attributes. A statistical segmentation was developed following which nonlinear differential price elasticity function was built. For each segment, price elasticity along with an optimization algorithm with complex constraints was developed to optimize price and achieve optimal revenue. – Pooja Dewan (committee chair of the Innovative Applications in Analytics Award) J A N U A R Y / F E B R U A R Y 2 014 | 29