Warner Optimizing the Display and Interpretation of Data

1. Auflage 2015
ISBN: 978-0-12-805341-6
Verlag: Elsevier Science & Techn.
Format: EPUB
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 148 Seiten

Reihe: Computer Science Reviews and Trends

'What information do these data reveal?' 'Is the information correct?' 'How can I make the best use of the information?' The widespread use of computers and our reliance on the data generated by them have made these questions increasingly common and important. Computerized data may be in either digital or analog form and may be relevant to a wide range of applications that include medical monitoring and diagnosis, scientific research, engineering, quality control, seismology, meteorology, political and economic analysis and business and personal financial applications. The sources of the data may be databases that have been developed for specific purposes or may be of more general interest and include those that are accessible on the Internet. In addition, the data may represent either single or multiple parameters. Examining data in its initial form is often very laborious and also makes it possible to 'miss the forest for the trees' by failing to notice patterns in the data that are not readily apparent. To address these problems, this monograph describes several accurate and efficient methods for displaying, reviewing and analyzing digital and analog data. The methods may be used either singly or in various combinations to maximize the value of the data to those for whom it is relevant. None of the methods requires special devices and each can be used on common platforms such as personal computers, tablets and smart phones. Also, each of the methods can be easily employed utilizing widely available off-the-shelf software. Using the methods does not require special expertise in computer science or technology, graphical design or statistical analysis. The usefulness and accuracy of all the described methods of data display, review and interpretation have been confirmed in multiple carefully performed studies using independent, objective endpoints. These studies and their results are described in the monograph. Because of their ease of use, accuracy and efficiency, the methods for displaying, reviewing and analyzing data described in this monograph can be highly useful to all who must work with computerized information and make decisions based upon it. - The reader will learn methods for easily increasing the speed and accuracy of reviewing data that are relevant to many diverse fields of endeavor - This will reduce the drudgery associated with reviewing the data and simultaneously improve the reliability of the interpretations that result from the review - This increased efficiency of review will make it easier to provide 'full disclosure' of the data to all those responsible for making decisions based on any actionable information that might be contained in the data

Robert A. Warner, MD is board-certified in both Internal Medicine and Cardiology and is currently living and working near Portland, Oregon. He received his BS degree from Union College in 1964 and his MD degree from Upstate Medical Center in 1969. He completed a residency in internal medicine at Upstate in 1972 and a fellowship in cardiology at Duke University Medical Center in 1975. He was an Eliphalet Nott Scholar at Union College and is a member of the Phi Beta Kappa, Sigma Xi and Alpha Omega Alpha honor societies. Dr. Warner was a member of the faculty of Upstate Medical Center College of Medicine from 1975 to 1998 where he rose to the rank of Full Professor of Medicine. From 1986 to 1996, he served as Chief of the Medical Service at the Syracuse VA Medical Center. From 1998 to 2002, Dr. Warner did medical research at the Duke University Clinical Research Institute in Durham, North Carolina and from 2002 to 2006 served as the Medical Director of Inovise Medical, Inc. in Portland, Oregon. Since then, he has remained active in research, continues to publish in medical and computer science journals and frequently presents his research findings at scientific meetings. Most of his current work consists of improving the accuracy of medical diagnoses and optimizing the interpretation of computer-generated data. The methods of interpretation that he has developed apply not only to biomedical data, but are also relevant to such diverse fields as engineering, the physical and social sciences, business and personal finance. Dr. Warner is the author of over 75 papers and 95 abstracts that have been published in peer-reviewed journals. He has served as consultant to many companies in the medical device and pharmaceutical industries and is the holder of four patents that are all related to the display and interpretation of biomedical data.

Autoren/Hrsg.

Warner, Robert

Leseproben

Chapter 1

The Scope and Importance of Computerized Data

Abstract

Many people currently have access to a great deal of computerized data that can serve many purposes. However, to make optimal use of this information, it is important to have methods of reviewing and analyzing this information in comprehensive, accurate, and efficient ways. An effective strategy for doing this is to use familiar computer programs to generate analog displays of available data that reveal useful and easily recognizable patterns in the information. Once relevant patterns in the information are recognized, the same digital data that were used to produce the analog displays can be used to study those patterns with great precision. In these ways, we can benefit greatly from what the availability of computerized data has to offer.

Keywords

data availability

analog and digital conversion

At the present time, computers generate enormous amounts of electronic data that are intended to serve the diverse needs of many people throughout the world. The availability of personal computers, in conjunction with access to the Internet, has helped many to use to these data to achieve their professional, commercial, and individual goals. Since computerized information can serve many important purposes, it is crucial that there be practical methods for using it most effectively. Features of such methods certainly include the need to interpret the data correctly. Also, the sheer volume of data available to us has also made it important that these methods enable us to review and analyze the data efficiently. Such reviews and analyses must be comprehensive. In other words, the speed with which one can review and analyze data should not come at the price of ignoring or discarding information that may be important.

Also, to be of greatest general use, our methods of displaying, reviewing, and analyzing data should be flexible and easy to employ. For example, a broadly applicable method that uses commonly available software is likely to be useful to more people than a method that requires the skills of computer programmers to accomplish only a highly specific task.

Optimal methods for displaying, reviewing, and interpreting computerized data can benefit people of many different interests and occupations. Such individuals include nurses, physicians, individual investors, businessmen, financial analysts, social scientists, engineers, economists, geologists, actuaries, political pollsters, and meteorologists.

The output of computers typically consists of digital data, i.e., numbers. Surely, there are plenty of different ways of analyzing numerical data. However, many of these methods, such as various types of mathematical modeling, generally require considerable training and skill in mathematics and statistics. In contrast, more intuitive ways of handling numerical data are available. For example, it is often advantageous to display the digital data not as numbers per se, but rather as pictorial images. We do this whenever we use the numbers to produce graphs of the data. In other words, we can often benefit from transforming the information provided by the computer from digital to analog form. For example, individual investors and investment professionals often employ “technical analysis” to try to predict whether the prices of stocks, bonds, mutual funds, and exchange-traded funds will rise or fall in the future. Technical analysis consists of studying previous patterns of variation in the prices of these securities. The investors and professional analysts hope that certain patterns of price fluctuation shown by these “technical charts” will help them buy securities when the prices are low and sell them when the prices are higher. Table 1.1 shows digital data that represent the closing prices in dollars of a security that mirrors the S&P 500 Index during the preceding 12-month period. Each number is the closing price in dollars of that security for each trading day of the previous year. The numbers are arranged in order from left to right and top to bottom, in the same fashion as the words on a page of English text. The information shown in Figure 1.1 is identical to that in Table 1.1, but is provided in analog, rather than in digital form. Figure 1.1 is simply a line graph of the digital data in Table 1.1 and shows, from left to right, the daily changes in the closing price of the security during the same 12-month period. Figure 1.1 is an example of a “technical chart.” The column of numbers on the left side of the graph represents the security’s closing price in dollars. Table 1.1 and Figure 1.1 contain identical information. However, the analog presentation of the information in Figure 1.1 makes it much easier to detect temporal patterns in the fluctuations of the security’s price than is possible by examining the digital data in Table 1.1.

Table 1.1

Digital Representation of a Security’s Closing Prices in Dollars

182.89

182.36

183.48

183.52

183.64

184.14

181.69

183.67

184.66

183.64

184.18

184.3

182.79

178.89

178.01

179.07

177.35

179.23

174.17

175.39

175.17

177.48

179.68

180.01

181.98

182.07

183.01

184.24

183.02

184.1

183.89

184.91

184.84

184.85

185.82

186.29

187.58

187.75

188.18

188.26

188.16

187.23

187.28

185.18

184.66

187.66

186.66

187.75

186.2

185.43

186.31

184.97

184.58

185.49

188.25

188.88

188.63

186.4

184.34

185.1

187.09

183.16

181.51

184.2

186.13

186.39

187.04

187.89

187.45

187.83

186.29

186.88

188.31

188.33

188.06

188.42

186.78

187.88

187.68

187.96

189.79

189.06

187.4

188.05

188.74

187.55

189.13

189.59

190.35

191.52

192.37

192.68

192.9

192.8

193.19

194.45

195.38

195.58

195.6

193.54

194.13

194.29

194.83

196.26

196.48

195.94

195.88

194.7

195.44

195.82

195.72

197.03

197.23

197.96

195.71

197.71

197.34

198.64

198.65

197.72

197.8

196.95

196.98

193.09

192.5

193.89

192.07

191.03

193.24

193.8

193.53

194.84

195.76

195.72

197.36

198.92

199.5

199.19

200.2

200.33

200.25

200.14

200.71

200.61

200.21

201.11

200.59

199.32

200.07

200.3

199.13

198.98

200.48

201.82

200.7

199.15

198.01

199.56

196.34

197.9

197.54

197.02

194.38

196.52

196.29

193.26

196.64

192.74

190.54

187.41

187.7

186.27

188.47

190.3

194.07

192.69

194.93

196.43

196.16

198.41

199.38

201.66

201.77

201.07

202.34

203.15

203.34

203.98

204.18

204.19

204.24

204.37

205.55

205.22

205.58

206.68

207.26

207.11

207.2

205.76

207.09

207.89

207.66

208

206.61

206.47

203.16

200.89

199.51

197.91

201.79

206.78

206.52

207.47

207.75

207.77

208.72

207.6

205.54

205.43

201.72

199.82

202.31

205.9

204.25

202.08

Fig. 1.1Analog representation of the data in Table 1.1.

Seismography is another field that demonstrates the value of presenting data in analog form. Geologists rely on seismographs...