Using Data and Text Mining to drive Innovation

 Text analytics - Using techniques such as Text Mining or Sentiment Analysis to exploit unstructured sources. These can range from verbatim adverse event descriptions to blog entries and tweets.  Data mining - Using advanced analytical techniques to identify patterns and associations in the data that indicate good or bad implications for the company.  Simulation – Using existing data to generate the parameters for simulations of clinical trials.  Forecasting – Predicting the most likely usage of clinical trial supplies. ANALYTICAL DECISION MAKING Decisions in life are frequently made on the basis of gut feel or instinct, and business decision making is no exception. This has many dangers, as what seems instinctively right may be completely misguided. For example, if you were able to fold a piece of paper in half forty times, what would you expect the combined thickness to be? Gut feel might tell you that the result would be a few inches or feet thick, whereas the correct answer for a piece of paper 0.193mm thick is 131,862 miles (or about halfway to the moon). The application of numerical methods to real world problems is the only way to eliminate bias in decision making and make full use of the available facts. This paper will cover some of the common techniques that can be applied to data available to the pharmaceutical industry and how this can take the industry forward. DATA MINING WHAT IS DATA MINING? Data Mining is the analysis of large volumes of data to uncover previously unsuspected patterns for business benefit. Data mining covers a wide variety of techniques such as Predictive modelling, Variable Selection, Association and Segmentation.