One of the most powerful steps organizations can take today to modernize their business intelligence (BI) capabilities is integrating predictive analytics. Predictive analytics is a practice that encompasses a variety of statistical techniques that analyze current and historical facts to make predictions about future or otherwise unknown events. It is becoming increasingly valuable in helping organizations uncover hidden trends, and taking that information a step further to develop forecasts or models for predicting how customers might respond to product or service offerings. In addition, new predictive analytics technologies enable self-service analytics for business users.
From Business Intelligence to Predictive Analytics
Business intelligence (Bl) is an umbrella term that covers architectures, tools, databases, analytical tools, applications, and methodologies. It means different things to different people. Part of the confusion about Bl lies in the range of acronyms and buzzwords associated with it and used within the industry. The visual below shows how it all fits together.
A BI system has four major components:
  • a data warehouse, with its source data
  • business and predictive analytics, a collection of tools for mining and analyzing the data in the data warehouse
  • business performance management (BPM) for monitoring and analyzing performance
  • a user interface, such as a dashboard
Data mining is defined as the process of discovering patterns in data. It is about solving problems by analyzing data already present in databases. Data mining is not a new discipline, but rather a new definition for the use of many disciplines. Data mining intersects many disciplines, including statistics, artificial intelligence, machine learning, management science, information systems, and databases.
By using existing and relevant data, data mining builds models to identify patterns among the attributes presented in the dataset. Models are the mathematical representations that identify the patterns among the attributes of the objects (e.g., customers) described in the data set. Some of these patterns are explanatory (explaining the interrelationships and affinities among the attributes), whereas others are predictive (foretelling future values of certain attributes).
In general, machine learning seeks to identify four major types of patterns:
  • Associations find the commonly co-occurring groupings of things, such as beer and diapers going together in the market-basket analysis.
  • Clusters identify natural groupings of things based on their known characteristics, such as assigning customers to different segments based on their demographics and past purchase behaviors.
  • Sequential relationships discover time-ordered events, such as predicting that an existing banking customer who already has a checking account will open a savings account followed by an investment account within a year.
  • Predictions tell the nature of future occurrences of certain events based on what has happened in the past, such as predicting the winner of the Super Bowl or forecasting the absolute temperature of a particular day.
Predictive Analytics
Predictive modeling is perhaps the most commonly practiced area of data mining and machine learning. It allows decision makers to estimate what the future holds through learning from the past.
The most popular predictive modeling techniques are artificial neural networks, support vector machines, and k-Nearest Neighbor.
An ANN is based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain. ANN systems “learn” to perform tasks by considering examples, generally without being programmed with any task-specific rules.
Support Vector Machines (SVMs) are supervised learning models with associated learning algorithms that analyze data used for classification and regression analysis. The goal of SVM is to generate mathematical functions that map input variables to desired outputs for classification- or regression-type prediction problems. They have a solid mathematical foundation, making them widely popular with data scientists.
k-Nearest Neighbor (k-NN) is a simplistic and logical prediction method that produces very competitive results. k-NN is a prediction method for classification as well as regression types. It is similar but significantly less computationally intensive than ANN and SVM, and also popular with data scientists, statisticians, and even business users due to its relative simplicity.
These techniques are capable of addressing both classification- and regression-type prediction problems. Often, they are applied to complex prediction problems where other methods are not capable of producing satisfactory results. Traditional statistical analysis or optimization techniques such as linear or quadratic programming may fall short in identifying certain patterns or solutions since they tend to focus on proving or disproving a hypothesis. Machine learning is used to find the right hypothesis and frame it in a computational model.
Tools vs. Solutions
A predictive analytics platform is typically operated by data scientists with the objective of creating models that can provide insights to key business decision makers. There are platforms of every size and cost to enable a modern analytics platform in an organization, from powerful commercial products like Datarobot to popular open-source solutions like Apache Spark and Python.
The key is to focus on solutions rather than on tools. At TESCHGlobal, we are tool-agnostic. We can support clients’ modern analytics strategies in a wide range of AI and machine learning platforms. Our focus is to assist customers in creating the right approach to develop models that will produce meaningful and actionable insights.
TG’s Value Proposition
Predictive analytics is critical in modernizing your data management strategy. We believe that empowering users with a platform capable of producing recommendations for actions based on insights generated from historical data will significantly improve your business processes.
Our predictive analytics specialists can enable your organization with a strategy that helps modernize your BI and data mining capabilities by implementing predictive modeling techniques that can be applied to a wide range of problem areas, from standard business problems of assessing customer needs to understanding and enhancing the efficiency of production processes to improving healthcare and medicine.
Sharda, Ramesh; Delen, Dursun; Turban, Efraim; Aronson, Janine; Liang, Ting Peng. Business Intelligence and Analytics: Systems for Decision Support (Page 15). Pearson Education, 2015.
Witten, Ian H., et al. Data Mining: Practical Machine Learning Tools and Techniques. Morgan Kaufmann, 2017.