Statistical Engine Clause Samples

Statistical Engine. In this section we will give an overview of the prototype statistical engine module. As mentioned earlier we aim at making our prototype as generic as possible while at the same time catering to the many different data scenarios identified (i.e. batch/streaming big/small data). Due to the diverse nature of the data involved, one of the major challenges we are likely to encounter is in enabling the statistical engine to efficiently communicate with the sensor database modules. This will require the sensor database to be able to store the data in a way which is optimal for the statistical process at hand. All the specifications for both the statistical engine and the modules in the IT stage are dictated both by the sensor networks characteristics and by the optimizer’s specifications. This will require interfaces facilitating communication between the optimizer and the statistical engine and between the statistical engine and the sensor database (see figure below). As part of our WP we plan on performing a research and development stage aimed at developing the optimal interfaces which should be generic enough to apply to all possible scenarios. The statistical engine should also be able to run on automation with minimal user interference. Figure 3: Interfaces in the Data Query Process
Statistical Engine. The purpose of the statistical engine is to analyze the data, which arrives from module 3, after or before (depending on the response time) it is stored in the database, in preparation for the optimizer’s queries. As part of our WP we will develop a prototype for a data interface between the statistical engines and the controller/optimizer (see Chapter 6 for further details).