F&R is a data-centric and computation heavy environment. Financial related data can be divided into two broad categories, i.e. pricing data and non-pricing data. Non-pricing data is becoming more and more important and its volume has grown significantly recently. Traditional financial analyses focus on price only without directly interpreting other sources of data. It has become essential that many firms start to look at non-pricing data to understand and provide explanation to what is happening in the market that causes the price movements. This requires large-scale intelligent analyses of non-pricing data, such as social media, usage logs, news, etc. At Thomson Reuters, we have three large areas of advanced analytics development for F&R, i.e. data science/analysis, natural language processing, and machine learning. For sentiment analysis, we consider news and social sentiment indicators. In this talk, I will discuss some of the efforts in these areas.