- Shopping Bag ( 0 items )
Combine business sense, statistics, and computers in a new and intuitive way, thanks to Big Data
Predictive analytics is a branch of data mining that helps predict probabilities and trends. Predictive Analytics For Dummies explores the power of predictive analytics and how you can use it to make valuable predictions for your business, or in fields such as advertising, fraud detection, politics, and others. This practical book does not bog you down with loads of mathematical or ...
Combine business sense, statistics, and computers in a new and intuitive way, thanks to Big Data
Predictive analytics is a branch of data mining that helps predict probabilities and trends. Predictive Analytics For Dummies explores the power of predictive analytics and how you can use it to make valuable predictions for your business, or in fields such as advertising, fraud detection, politics, and others. This practical book does not bog you down with loads of mathematical or scientific theory, but instead helps you quickly see how to use the right algorithms and tools to collect and analyze data and apply it to make predictions.
Topics include using structured and unstructured data, building models, creating a predictive analysis roadmap, setting realistic goals, budgeting, and much more.
Propose, produce, and protect predictive analytics projects through your company with Predictive Analytics For Dummies.
Entering the Arena
In This Chapter
* Explaining the building blocks
* Probing capabilities
* Surveying the market
Predictive analytics is a bright light bulb powered by your data.
You can never have too much insight. The more you see, the better the decisions you make — and you never want to be in the dark. You want to see what lies ahead, preferably before others do. It's like playing the game "Let's Make a Deal" where you have to choose the door with the hidden prize. Which door do you choose? Door 1, Door 2, or Door 3? They all look the same, so it's just your best guess — your choice depends on you and your luck. But what if you had an edge — the ability to see through the keyhole? Predictive analytics can give you that edge.
Exploring Predictive Analytics
What would you do in a world where you know how likely you are to end up marrying your college roommate? Where you can predict what profession will best suit you? Where you can predict the best city and country for you to live in?
In short, imagine a world where you can maximize the potential of every moment of your life. Such a life would be productive, efficient, and powerful. You will (in effect) have superpowers — and a lot more spare time. Well, such a world may seem a little boring to people who like to take uncalculated risks, but not to a profit-generating organization. Organizations spend millions of dollars managing risk. And if there is something out there that helps them manage their risk, optimize their operations, and maximize their profits, you should definitely learn about it. That is the world of predictive analytics.
Big data is the new reality. In fact, data is only getting bigger, faster, and richer. It's here to stay and you'd better capitalize on it.
Data is one of your organization's most valuable assets. It's full of hidden value, but you have to dig for it. Data mining is the discovery of hidden patterns of data through machine learning — and sophisticated algorithms are the mining tools. Predictive analytics is the process of refining that data resource, using business knowledge to extract hidden value from those newly discovered patterns.
Data mining + business knowledge = predictive analytics = value
Today's leading organizations are looking at their data, examining it, and processing it to search for ways to better understand their customer base, improve their operations, outperform their competitors, and better position themselves in the marketplace. They are looking into how they can use that information to increase their market share and sharpen their competitive edge. How can they drive better sales and more effectively targeted marketing campaigns? How can they better serve their customers and meet their needs? What can they do to improve the bottom line?
But these tools are useful in realms beyond business. As one major example, government law enforcement agencies are asking questions related to crime detection and prevention. Is this a person of interest? Is this person about to commit a heinous crime? Will this criminal be a repeat offender? Where will the next crime happen?
Other industries, notably those with financial responsibility, could use a trustworthy glimpse into the future. Companies are trying to know ahead of time whether the transaction they're currently processing is fraudulent, whether an insurance claim is legitimate, whether a credit card purchase is valid, whether a credit applicant is worthy of credit ... the list goes on.
Governments, companies, and individuals are (variously) looking to spot trends in social movements, detect emerging healthcare issues and disease outbreaks, uncover new fashion trends, or find that perfect lifetime partner.
These — and plenty more — business and research questions are topics you can investigate further to find answers to by mining the available data and building predictive analytics models to guide future decisions.
Data + predictive analytics = light.
Highlighting the model
A model is a mathematical representation of an object or a process. We build models to simulate real-world phenomena as a further investigative step, in hopes of understanding more clearly what's really going on. For example, to model our customers' behavior, we seek to mimic how our customers have been navigating through our websites:
[check] What products did they look at before they made a purchase?
[check] What pages did they view before making that purchase?
[check] Did they look at the products' descriptions?
[check] Did they read users' reviews?
[check] How many reviews did they read?
[check] Did they read both positive and negative reviews?
[check] Did they purchase something else in addition to the product they came looking for?
We collect all that data from past occurrences. We look at those historical transactions between our company and our customers — and try to make consistent sense of them. We examine that data and see whether it holds answers to our questions. Collecting that data — with particular attention to the breadth and depth of the data, its quality level, and its predictive value — helps to form the boundaries that will define our model and its outputs.
This process is not to be confused with just reporting on the data; it's also different from just visualizing that data. Although those steps are vital, they're just the beginning of exploring the data and gaining a usable understanding of it.
We go a lot deeper when we're talking about developing predictive analytics. In the first place, we need to take a threefold approach:
[check] Thoroughly understand the business problem we're trying to solve.
[check] Obtain and prepare the data we want our model to work with.
[check] Run statistical analysis, data-mining, and machine-learning algorithms on the data.
In the process, we have to look at various attributes — data points we think are relevant to our analysis. We'll run several algorithms, which are sets of mathematical instructions that get machines to do problem-solving.
We keep running through possible combinations of data and investigate what-if scenarios. Eventually we find our answers, build our model, and prepare to deploy it and reap its benefits.
What does a model look like? Well, in programming terms, a predictive analytics model can be as simple as a few if ... then statements that tell the machine, "If this condition exists, then perform this action."
Here are some simple rule-based trading models:
[check] If it's past 10:00a.m. ET and the market is up, then buy 100 shares of XYZ stock.
[check] If my stock is up by 10 percent, then take profits.
[check] If my portfolio is down by 10 percent, then exit my positions.
Here's a simple rule-based recommender system (for more about recommender systems, see Chapter 2):
[check] If a person buys a book by this author, then recommend other books by the same author.
[check] If a person buys a book on this topic, then recommend other books on the same and related topic.
[check] If a person buys a book on this topic, then recommend books that other customers have purchased when they bought this book.
Adding Business Value
In an increasingly competitive environment, organizations always need ways to become more competitive. Predictive analytics found its way into organizations as one such tool. Using technology in the form of machine-learning algorithms, statistics, and data-mining techniques, organizations can uncover hidden patterns and trends in their data that can aid in operations and strategy and help fulfill critical business needs.
Embedding predictive analytics in operational decisions improves return on investment because organizations spend less time dealing with low-impact, low-risk operational decisions. Employees can focus more of their time on high-impact, high-risk decisions. For instance, most standard insurance claims can be automatically paid out. However, if the predictive model comes across a claim that's unusual (an outlier), or if the claim exhibits the same pattern as a fraudulent claim, the system can flag the claim automatically and send it to the appropriate person to take action.
By using predictive analytics to predict a future event or trend, the company can create a strategy to position itself to take advantage of that insight. If your predictive model is telling you (for example) that the trend in fashion is toward black turtlenecks, you can take appropriate actions to design more black-colored turtlenecks or design more accessories to go with the fashionable item.
Organizations around the world are striving to improve, compete, and be lean. They're looking to make their planning process more agile. They're investigating how to manage inventories and optimize the allocations of their human resources to best advantage. They're looking to act on opportunities as they arise in real time.
Predictive analytics can make all those goals more reachable. The domains to which predictive analytics can be applied are unlimited; the arena is wide open and everything is fair game. Let the mining start. Let the analysis begin.
Go to your analytics team and have them mine the data you've accumulated or acquired, with an eye toward finding an advantageous niche market for your product; innovate with data. Ask the team to help you gain confidence in your decision-making and risk management.
Albert Einstein once said, "Know where to find information and how to use it; that is the secret of success." If that's the secret to success, then you will succeed by using predictive analytics: The information is in your data and data mining will find it. The rest of the equation relies on your business knowledge of how to interpret that information — and ultimately use it to create success.
Finding value in data equals success. Therefore we can rewrite our predictive analytics equation as
Data mining + business knowledge = predictive analytics = success
Empowering your organization
Predictive analytics empowers your organization by providing three advantages:
Predictive analytics will lead you to see what is invisible to others — in particular, useful patterns in your data.
Predictive analytics can provide you with powerful hints to lend direction to the decisions you're about to make in your company's quest to retain customers, attract more customers, and maximize profits. Predictive analytics can go through a lot of past customer data, associate it with other pieces of data, and assemble all the pieces in the right order to solve that puzzle in various ways, including
[check] Categorizing your customers and speculate about their needs.
[check] Knowing your customers' wish lists.
[check] Guessing your customers' next actions.
[check] Categorizing your customers as loyal, seasonal, or wandering.
Knowing this type of information beforehand shapes your strategic planning and helps optimize resource allocation, increase customer satisfaction, and maximize your profits.
A well-made predictive analytics model provides analytical results free of emotion and bias. The model uses mathematical functions to derive forward insights from numbers and text that describe past facts and current information. The model provides you with consistent and unbiased insights to support your decisions.
Consider the scenario of a typical application for a credit card: The process takes a few minutes; the bank or agency makes a quick, fact-based decision on whether to extend credit, and is confident in their decision. The speed of that transaction is possible thanks to predictive analytics, which predicted the applicant's creditworthiness.
Imagine having to read a lot of reports, derive insights from the past facts buried in them, go through rows of Excel spreadsheets to compare results, or extract information from a large array of numbers. You'd need a staff to do these time-consuming tasks. With predictive analytics, you can use automated tools to do the job for you — saving time and resources, reduces human error, and improves precision.
For example, you can focus targeted marketing campaigns by examining the data you have about your customers, their demographics, and their purchases. When you know precisely which customers you should market to, you can zero in on those most likely to buy.
Starting a Predictive Analytic Project
For the moment, let's forget about algorithms and higher math; predictions are used in every aspect of our lives. Consider how many times you have said (or heard people say), "I told you that was going to happen."
If you want to predict a future event with any accuracy, however, you'll need to know the past and understand the current situation. Doing so entails several processes:
[check] Extract the facts that are currently happening.
[check] Distinguish present facts from those that just happened.
[check] Derive possible scenarios that could happen.
[check] Rank the scenarios according to how likely they are to happen.
Predictive analytics can help you with each of these processes, so that you know as much as you can about what has happened and can make better-informed decisions about the future.
Companies typically create predictive analytics solutions by combining three ingredients:
[check] Business knowledge
[check] Data-science team and technology
[check] The data
Though the proportion of the three ingredients will vary from one business to the next, all are required for a successful predictive analytic solution that yields actionable insights.
Because any predictive analytics project is started to fulfill a business need, business-specific knowledge and a clear business objective are critical to its success. Ideas for a project can come from anyone within the organization, but it's up to the leadership team to set the business goals and get buy-in from the needed departments across the whole organization.
Be sure the decision-makers in your team are prepared to act. When you present a prototype of your project, it needs an in-house champion — someone who's going to push for its adoption.
The leadership team or domain experts must also set clear metrics — ways to quantify and measure the outcome of the project. Appropriate metrics keep the departments involved are clear about what they need to do, how much they need to do, and whether what they're doing is helping the company achieve its business goals.
The business stakeholders are those who are most familiar with the domain of the business. They'll have ideas about which correlations — relationships between features — of data work and which don't, which variables are important to the model, and whether you should create new variables — as in derived features or attributes — to improve the model.
Business analysts and other domain experts can analyze and interpret the patterns discovered by the machines, making useful meaning out of the data patterns and deriving actionable insights.
This is an iterative (building a model and interpreting its findings) process between business and science. In the course of building a predictive model, you have to try successive versions of the model to improve how it works (which is what data experts mean when they say iterate the model over its lifecycle). You might go through a lot of revisions and repetitions before you can prove that your model is bringing real value to the business. Even after the predictive models are deployed, the business must monitor the results, validate the accuracy of the models and improve upon the models as more data is being collected.
Excerpted from Predictive Analytics For Dummies by Anasse Bari, Mohamed Chaouchi, Tommy Jung. Copyright © 2014 John Wiley & Sons, Ltd. Excerpted by permission of John Wiley & Sons.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Part I: Getting Started with Predictive Analytics 5
Chapter 1: Entering the Arena 7
Chapter 2: Predictive Analytics in the Wild 19
Chapter 3: Exploring Your Data Types and Associated Techniques 43
Chapter 4: Complexities of Data 57
Part II: Incorporating Algorithms in Your Models 73
Chapter 5: Applying Models 75
Chapter 6: Identifying Similarities in Data 89
Chapter 7: Predicting the Future Using Data Classification 115
Part III: Developing a Roadmap 145
Chapter 8: Convincing Your Management to Adopt Predictive Analytics 147
Chapter 9: Preparing Data 167
Chapter 10: Building a Predictive Model 177
Chapter 11: Visualization of Analytical Results 189
Part IV: Programming Predictive Analytics 205
Chapter 12: Creating Basic Prediction Examples 207
Chapter 13: Creating Basic Examples of Unsupervised Predictions 233
Chapter 14: Predictive Modeling with R 249
Chapter 15: Avoiding Analysis Traps 275
Chapter 16: Targeting Big Data 295
Part V: The Part of Tens 307
Chapter 17: Ten Reasons to Implement Predictive Analytics 309
Chapter 18: Ten Steps to Build a Predictive Analytic Model 319
Posted March 23, 2014
Posted August 19, 2014