Building products that delight customers requires insights from multiple data channels: usage, social sentiment, and direct interactions. At Salesforce.com, the Product Intelligence team evaluates all of these channels to provide the best predictions of product adoption. We boil down all of the data and measure successful product adoption along two dimensions: usage and sentiment. The latter is important because it differentiates between our best adopters and those that are feeling trapped. The insights provided by Product Intelligence lead to better products that result in high adoption and low attrition.

Large scale user generated data sets have a huge potential for creating insights that can be embedded into products. However, to spot patterns and extract useful information from massive data is hard. Even when computational capacity is not an issue, the size, the noise and the lack of well defined problems poses huge challenges. In this talk we will explore the challenges and opportunities in this space using examples from the work I have done with YouTube, Google Plus and Google TV Ads.

I will first share some of the techniques I use to eliminate confusion in ill-defined problems. I will then use examples from business and science to illustrate how insight emerges from a multivariate analysis. I will conclude with a discussion of scale and its impact on selecting one’s data mining approach.

The Healthcare industry is at a turning point. Just as several years ago when quants moved from Wall Street into consumer internet to work on computational advertising and recommendation engines, the flow of analytical talent into the healthcare industry is currently on fire. This trend is driven by the realization that the best way to help physicians make better treatment decisions while decreasing cost is by leveraging data and predictive modeling. The economic opportunities are massive (the industry is approaching 20% of US GDP), the ability to have massive societal impact is clear, and the large datasets of hundreds of millions of patient encounters grow significantly with the addition of genomic data. With data and data scientists at the core, healthcare is working to solve its many problems and move into an era of personalized medicine and sustainable cost growth.

- The Past

- The Present

- The Future

• How to Build a Data Science Team- 

• The Challenges of Big Data

• Open Data

• Managing Your Data

(a) Workforce Development for Data Science.

(b) What are the fundamental skills of the Data Scientist?

(c) What topics come under the heading "Data Science"?

(d) Data Science in Education at all age levels: how, what, why?

(e) What are lessons learned from Data Science practice?

(f) What are current challenges in Data Science advancement?