Predictive Analytics capabilities are provided using open source solutions in the two languages R and Python, as well as RStat module, and Web Focus Reporting system. Various algorithm types can be used including regression, classifications, recommender systems, clustering, preprocessing, model selection, etc. Predictive Analytics is provided via Splunk as well, e.g. detection of abnormal activities using log data. Predictive Analytics answers the question "what is probably going to happen in the future?", while Descriptive Analytics looks at historical data to answer the question "what happened in the past with the business?"
Prescriptive analytics provides the best outcomes for modeled situations based on optimization. Optimization entails deciding how to best leverage limited assets, time and resources in situations with varying levels of uncertainty. Prescriptive analytics tries to answer the question: "Now what?", giving recommendations for key decisions based on future outcomes. Prescriptive analytics capabilities are provided based on the large availability of open source solutions. An example is exam-scheduling optimization that minimizes exam conflicts.
Big Data Analytics & Management
Big Data capabilities are provided via Apache Spark on on HPC (High-Performance Computing Cluster). It is the application of "supercomputers" to computational problems that are either too large for standard computers or would take too long. Machine Learning is provided via MLlib library.
Natural Language Processing (NLP)
NLP techniques and technologies can enhance educational systems and applications. Students may find the best answers in coursework data; Workload may be reduced for staff reviewing thousands of applications as an NLP model may summarize large content.
Cloud resources are currently and selectively used in our projects (Azure). Other cloud solutions (AWS, GCP) are currently used for academic purposes.
Statistical Reporting is involved when population data is unavailable. Based on sample data, such as surveys, inferences can be made about the population.