English | MP4 | AVC 1280×720 | AAC 44KHz 2ch | 1h 54m | 468 MB
Are you a data science practitioner, looking to develop or enhance your skills in predictive analysis and data mining? This course provides several “big picture” insights, via instructor Keith McCormick, a veteran practitioner who has completed dozens of real-world projects. Keith begins by introducing you to key definitions and processes that you will need to complete the course successfully. He steps you through defining the problem you need your predictive analysis to address, then focuses on how to make sure you meet the data requirements and how good data preparation improves your data mining projects. Keith dives into the skill sets and resources that you need and the problems you will face. Then he goes over the steps to find the solution and put it to work with probabilities, propensities, missing data, meta modeling, and much more. Keith finishes up with detailed explanations of CRISP-DM and Tom Khabaza’s nine laws of data mining, plus Tom’s new 10th law.
Table of Contents
Introduction
1 Data mining and predictive analytics
1. What Is Data Mining and Predictive Analytics
2 Introducing the essential elements
3 Defining data mining
4 Introducing CRISP-DM
2. Problem Definition
5 Beginning with a solid first step Problem definition
6 Framing the problem in terms of a micro-decision
7 Why every model needs an effective intervention strategy
8 Evaluate a project’s potential with business metrics and ROI
9 Translating business problems into data mining problems
3. Data Requirements
10 Understanding data requirements
11 Gathering historical data
12 Meeting the flat file requirement
13 Determining your target variable
14 Selecting relevant data
15 Hints on effective data integration
16 Understanding feature engineering
17 Developing your craft
4. Resources You Will Need
18 Skill sets and resources that you’ll need
19 Compare machine learning and statistics
20 Assessing team requirements
21 Budgeting sufficient time
22 Working with subject matter experts
5. Problems You Will Face
23 Anticipating project challenges
24 Addressing missing data
25 Addressing organizational resistance
26 Addressing models that degrade
6. Finding the Solution
27 Preparing for the modeling phase tasks
28 Searching for optimal solutions
29 Seeking surprise results
30 Establishing proof that the model works
31 Embracing a trial and error approach
7. Putting the Solution to Work
32 Preparing for the deployment phase
33 Using probabilities and propensities
34 Understanding meta modeling
35 Understanding reproducibility
36 Preparing for model deployment
37 How to approach project documentation
8. The Nine Laws of Data Mining
38 Understanding why models change
39 CRISP-DM and the laws of data mining
40 Understanding CRISP-DM
41 Advice for using CRISP-DM
42 Understanding the nine laws of data mining
43 Understanding the first and second laws
44 Understanding the data preparation law
45 Understanding the laws about patterns
46 Understanding the insight and prediction laws
47 Understanding the value law
Conclusion
48 Next steps
Resolve the captcha to access the links!