It's All Analytics - Part II: Designing an Integrated Ai, Analytics, and Data Science Architecture for Your Organization
暫譯: 一切皆分析 - 第二部分：為您的組織設計整合的 AI、分析與數據科學架構

企業的分析工作高達70%甚至更多都失敗了！！！即使這些企業在開發他們認為良好的數據和分析計劃上投入了大量的時間、人才和金錢，為什麼會這樣？因為高層主管、決策者以及整個分析團隊並未考慮到使這些分析工作成功的最重要的方面。在《一切都是分析！》系列的第二本書中，我們描述了兩個主要內容：1) 這個最重要的方面包含什麼，以及 2) 如何將這個最重要的方面置於分析工作的核心，從而使您的分析計劃成功。

本系列的第二本書分為三個主要部分：

第一部分，成功的組織設計，討論 ....... 整個公司及其分析團隊需要完全對齊，以使其分析工作成功。這意味著要關注文化——公司文化！！！要成功，公司的CEO和決策者必須充分意識到文化重點在於「建立分析卓越中心」。簡而言之，文化——公司文化是成功分析計劃中最重要的方面。重點必須放在創新上，因為這是分析團隊開發成功算法所需的，這將導致公司效率的提高和利潤的增加。

第二部分，成功的數據設計，討論 ..... 數據是分析成功的基石。您可以擁有最好的分析算法和模型，但如果沒有良好的數據，努力最多也只是平庸，甚至可能完全失敗。第二部分還進一步探討數據，描述了如揮發性數據存儲和非揮發性數據存儲等內容，還有數據結構和數據格式，並考慮了集群計算、數據沼澤、混亂數據、數據市集、企業數據倉庫、數據水庫和分析沙盒，以及數據虛擬化、策劃數據、購買數據、初生和未來數據、補充數據、有意義的數據、地理信息系統（GIS）和地理分析數據、圖形數據庫和時間序列數據庫。第二部分還考慮了數據治理，包括數據完整性、數據安全性、數據一致性、數據信心、數據洩漏、數據分佈和數據素養。

第三部分，分析技術設計以達成成功，討論 .... 分析成熟度及其各個方面，如探索性數據分析、數據準備、特徵工程、模型構建、模型評估、模型選擇和模型部署。第三部分還深入探討現代預測分析的細節，討論如AI = 人工智慧、機器學習、深度學習等術語，以及傳統分析中對現代分析有幫助的統計學、預測、優化和模擬等方面。第三部分還探討如何溝通和行動於分析，包括在您的公司/組織內建立成功的分析文化。

總而言之，如果您的公司或組織需要成功地使用分析，本書將為您提供實現這一目標所需的基本知識。

Scott Burk has been solving complex business and health care problems for twenty-five years through science, statistics, machine learning and business acumen. Scott started his career, well actually in analytics, as as an analytic chemist after graduating with a double major in biology and chemistry from Texas State University. He continued his education, going to school at night taking advanced courses in science and math at the University of Texas at Dallas (UTD). He then started programming at the toxicology lab where he was working and thus started taking computer science (CS) and business courses until he graduated with a Master's in Business with a concentration in finance soon after from UTD.

Texas Instruments (TI) hired him as a financial systems analyst in Semiconductor Group, but due to TI's needs and Scott's love of computers, he soon after became a systems analyst for corporate TI. He worked there for three years and started itching to get back to school (even though, he continued to take courses at night (Operations Research and CS) through TI's generous educational program). TI granted him an educational leave of absence and he went to Baylor University to teach in the business school and get a PhD in statistics. He joined Baylor as a non-tenure track professor teaching Quantitative Business Analysis (today = business analytics).

After graduating, Scott went back to TI as a Decision Support Manager for the consumer arm of TI (today = consulting data scientist). Where he engaged in many functional areas - marketing and sales, finance, engineering, logistics, customer relations the call center and more. It was a dream job, but unfortunately, TI exited that business.

Scott joined Scott and White, a large integrated healthcare delivery system in Texas as a consulting statistician. He moved into an executive role as Associate Executive Director, Information Systems leading Data Warehousing, Business Intelligence and Quality Organizations working with clinics, hospitals and the health plan. At the same time, he received a faculty appointment and taught informatics with Texas A&M University. He left, but later came back to Baylor, Scott and White (BSW) as Chief Statistician for BSW Healthplan.

Scott continued his education, getting an advanced management certification from Southern Methodist University (SMU) and Master's Degree (MS) in Data Mining (machine learning) from Central Connecticut State University. Scott is a firm believer in life-long learning.

He also worked as Chief Statistician at Overstock, re-engineering the way they tested and evaluated marketing campaigns and other programs (analytics, statistics). He launched their 'total customer value' program. He was a Lead Pricing Scientist (analytics, optimization) for a B2B pricing optimization company (Zilliant) for a number of years. He thoroughly enjoyed working with a rich diverse, well-educated group that affected the way he looks at multidisciplinary methods of solving problems.

He was a Risk Manager for eBay/Paypal identifying fraud and other risks on the platform and payment system. He has been working the last few years supporting software development, marketing and sales, specifically data infrastructure, data science and analytics platforms for Dell and now TIBCO. He supports his desire to learn and keep current by writing and teaching in the Masters of Data Science Program at City University of New York.

David Sweenor is an analytics thought leader, international speaker, author, and has co-developed several patents. David has over 20 years of hands-on business analytics experience spanning product marketing, strategy, product development, and data warehousing. He specializes in artificial intelligence, machine learning, data science, business intelligence, the internet of things (IoT), and manufacturing analytics.

In his current role as the Sr. Director of Product Marketing at Alteryx, David is responsible for GTM strategy for the data science and machine learning portfolio. Prior to joining Alteryx, David has served in a variety of roles--including an Analytics Center of Competency solutions consultant, competitive intelligence analyst, semiconductor yield characterization engineer, and various advanced analytics roles for SAS, IBM, TIBCO, Dell, and Quest. David holds a B.S. in Applied Physics from Rensselaer Polytechnic Institute in Troy, NY and an MBA from the University of Vermont.

Follow David on Twitter @DavidSweenor and connect with him on LinkedIn https: //www.linkedin.com/in/davidsweenor/.

Dr. Gary Miner received his B.S. from Hamline University, St. Paul, Minnesota with biology, chemistry and education majors; M.S. in Zoology & Population Genetics from the University of Wyoming, and his Ph.D. in Biochemical Genetics from the University of Kansas as the recipient of a NASA Pre-Doctoral Fellowship. During the doctoral study years, he also studied mammalian genetics at The Jackson Laboratory, Bar Harbor, ME, under a College Training Program on an NIH award; and another College Training Program at the Bermuda Biological Station, St. George's West, Bermuda in a Marine Developmental Embryology Course, on an NSF award; and a third College Training Program held at the University of California, San Diego at the Molecular Techniques in Developmental Biology Institute, again on an NSF award.

Following that he studied as a Post-Doctoral student at the University of Minnesota in Behavioral Genetics, where, along with research in schizophrenia and Alzheimer's Disease, he learned how to write books from assisting in editing two book manuscripts of his mentor, Irving Gottesman, Ph.D. (Dr. Gottesman returned the favor 41 years later by writing two tutorials for this PRACTICAL TEXT MINING book). After academic research and teaching positions, Dr. Miner did another two-year NIH-Post-Doctoral in Psychiatric Epidemiology and Biostatistics at the University of Iowa where he became thoroughly immersed in studying affective disorders and Alzheimer's Disease. All together he spend over 30 years researching and writing papers and books on the genetics of Alzheimer's Disease (Miner, G.D., Richter, R, Blass, J.P., Valentine, J.L, and Winters-Miner, Linda. FAMILIAL ALZHEIMER'S DISEASE: Molecular Genetics and Clinical Perspectives. Dekker: NYC, 1989; and Miner, G.D., Winters-Miner, Linda, Blass, J.P., Richter, R, and Valentine, J.L. CARING FOR ALZHEIMER'S PATIENTS: A Guide for Family & Healthcare Providers. Plenum Press Insight Books: NYC. 1989).

Over the years he held positions, including professor and chairman of a department, at various universities including The University of Kansas, The University of Minnesota, Northwest Nazarene University, Eastern Nazarene University, Southern Nazarene University, Oral Roberts University Medical School where he was Associate Professor of Pharmacology and Director of the Alzheimer Disease & Geriatric Disorders Research Laboratories, and even for a period of time in the 1990's was a visiting Clinical Professor of Psychology for Geriatrics at the Fuller Graduate School of Psychology & Fuller Theological Seminary in Pasadena, CA.

In 1985 he and his wife, Dr. Linda Winters-Miner [author of several tutorials in this book] founded The Familial Alzheimer's Disease Research Foundation [aka The Alzheimer's Foundation] which became a leading force in organizing both local and international scientific meetings and thus bringing together all the leaders in the field of genetics of AD from several countries, which then lead to the writing of the first scientific book on the genetics of Alzheimer's Disease; this book included papers by over 100 scientists coming out of the First International Symposium on the Genetics of Alzheimer's Disease held in Tulsa, OK in October, 1987. During part of this time he was also an Affiliate Research Scientist with the Oklahoma Medical Research Foundation located in Oklahoma City with the University of Oklahoma School of Medicine.

Dr. Miner was influential in bringing all of the world's leading scientists working on Genetics of AD together at just the right time when various laboratories from Harvard to Duke University and University of California-San Diego, to the University of Heidelberg, in Germany, and universities in Belgium, France, England and Perth, Australia were beginning to find genes which they thought were related to Alzheimer's Disease.

During the 1990's Dr. Miner was appointed to the Oklahoma Governor's Task Force on Alzheimer's Disease, and also Associate Editor for Alzheimer's Disease for THE JOURNAL OF GERIATRIC PSYCHIATRY & NEUROLOGY, which he still serves on to this day. By 1995 most of these dominantly inherited genes for AD had been discovered, and the one that Dr. Miner had been working on since the mid-1980's with the University of Washington in Seattle was the last of these initial 5 to be identified, this gene on Chromosome 1 of the human genome. At that time, having met the goal of finding out some of the genetics of AD, Dr. Miner decided to do something different, to find an area of the business world, and since he had been analyzing data for over 30 years, working for StatSoft, Inc. as a Senior Statistician and Data Mining Consultant seemed a perfect semi-retirement career. Interestingly (as his wife had predicted), he discovered that the business world was much more fun than the academic world, and at a KDD-Data Mining meeting in 1999 in San Francisco, he decided that he would specialize in data mining. Incidentally, he first met Bob Nisbet there who told him, You just have to meet this bright young rising star John Elder!, and within minutes Bob found John introduced me to him, as he was also at this meeting.

As Gary delved into this new data mining field, and looked at statistics text books in general, he saw the need for 'practical statistical books' and started writing chapters, and organizing various outlines for different books. Gary, Bob, and John kept running into each other at KDD meetings, and eventually at a breakfast meeting in Seattle in August of 2005 decided they needed to write a book on data mining, and right there re-organized Gary's outline which eventually became the book Handbook of Statistical Analysis and Data Mining Applications, 2009, published by Elsevier. And then, in 2012, he was the lead author on a 2nd book from Elsevier/Academic Press, PRACTICAL TEXT MINING. And then a 3rd in this series in 2015: PRACTICAL PREDICTIVE ANALYTICS and DECISIONING SYSTEMS FOR MEDICINE. All thanks to Dr. Irving Gottesman, Gary's mentor in book writing, who planted the seed back in 1970 while Gary was doing a post-doctoral with him at the University of Minnesota.

His latest book was released in 2018, the 2nd Edition of the 2009 book HANDBOOK OF STATISTICAL ANALYSIS and DATA MINING APPLICATIONS (https: //www.amazon.com/Handbook-Statistical-Analysis-Mining-Applications/dp/0124166326/); and a 2019 book written more for the layperson and decision maker, titled: HEALTHCARE'S OUT SICK - PREDIDCTING A CURE - SOLUTIONS THAT WORK!!! Published by Routledge / Taylor and Francis Group - A Productivity Press Book (https: //www.amazon.com/HEALTHCAREs-OUT-SICK-PREDICTING-INNOVATIONS/dp/1138581097).

Dr. Miner is currently working on a 2nd and 3rd book in a series with Scott Burk, Ph.D., and also teaches courses periodically in Predictive Analytics and Healthcare Analytics for the University of California-Irvine.

斯科特·伯克（Scott Burk）在商業和醫療保健領域解決複雜問題已有二十五年，運用科學、統計學、機器學習和商業智慧。斯科特的職業生涯起步於分析領域，作為一名分析化學家，他在德州州立大學獲得生物學和化學的雙主修學位後開始工作。他繼續進修，晚上在德克薩斯大學達拉斯分校（UTD）修讀科學和數學的進階課程。隨後，他在毒理學實驗室開始編程，並開始修讀計算機科學（CS）和商業課程，直到不久後從UTD獲得金融專業的商業碩士學位。

德州儀器（Texas Instruments, TI）聘請他擔任半導體部門的財務系統分析師，但由於TI的需求和斯科特對計算機的熱愛，他不久後成為TI公司的系統分析師。他在那裡工作了三年，開始渴望重返學校（儘管他繼續在晚上修讀TI慷慨的教育計劃下的運籌學和計算機科學課程）。TI批准了他的教育休假，他前往貝勒大學（Baylor University）在商學院任教並獲得統計學的博士學位。他以非終身職教授的身份加入貝勒大學，教授定量商業分析（今天稱為商業分析）。

畢業後，斯科特回到TI，擔任消費者部門的決策支持經理（今天稱為諮詢數據科學家）。他參與了許多功能領域——市場營銷和銷售、財務、工程、物流、客戶關係、呼叫中心等。這是一份夢想工作，但不幸的是，TI退出了該業務。

斯科特加入了斯科特與懷特（Scott and White），這是一個大型綜合醫療服務系統，擔任諮詢統計師。他轉任信息系統副執行董事，負責數據倉儲、商業智能和質量組織，與診所、醫院和健康計劃合作。同時，他獲得了教職，並在德州農工大學（Texas A&M University）教授信息學。他離開了，但後來又回到貝勒大學和斯科特與懷特（BSW），擔任BSW健康計劃的首席統計師。

斯科特繼續進修，獲得南方衛理公會大學（Southern Methodist University, SMU）的高級管理認證，並在中央康乃狄克州立大學獲得數據挖掘（機器學習）的碩士學位。斯科特堅信終身學習的重要性。

他還曾擔任Overstock的首席統計師，重新設計他們測試和評估市場營銷活動及其他計劃的方式（分析、統計）。他啟動了他們的“總客戶價值”計劃。他曾擔任B2B定價優化公司Zilliant的首席定價科學家（分析、優化）多年。他非常享受與一群多元化且受過良好教育的團隊合作，這影響了他看待解決問題的多學科方法。

他曾擔任eBay/Paypal的風險經理，識別平台和支付系統上的詐騙及其他風險。在過去幾年中，他一直支持軟件開發、市場營銷和銷售，特別是為Dell和現在的TIBCO提供數據基礎設施、數據科學和分析平台。他通過在紐約市立大學的數據科學碩士課程中寫作和教學來支持他學習和保持最新的願望。

大衛·斯維諾（David Sweenor）是一位分析領域的思想領袖、國際演講者、作者，並共同開發了幾項專利。大衛擁有超過20年的實際商業分析經驗，涵蓋產品市場營銷、策略、產品開發和數據倉儲。他專注於人工智慧、機器學習、數據科學、商業智能、物聯網（IoT）和製造分析。

在他目前擔任Alteryx的產品市場高級總監的角色中，大衛負責數據科學和機器學習產品組合的市場進入策略。在加入Alteryx之前，大衛曾擔任多種職位，包括分析能力中心的解決方案顧問、競爭情報分析師、半導體產量特徵工程師，以及在SAS、IBM、TIBCO、Dell和Quest的各種高級分析角色。大衛擁有來自紐約州特洛伊的倫斯勒理工學院的應用物理學學士學位，以及來自佛蒙特大學的MBA學位。

在推特上關注大衛@DavidSweenor，並在LinkedIn上與他聯繫 https://www.linkedin.com/in/davidsweenor/。

加里·邁納博士（Dr. Gary Miner）在明尼蘇達州聖保羅的哈姆萊大學獲得生物學、化學和教育的學士學位；在懷俄明大學獲得動物學和種群遺傳學的碩士學位；並在堪薩斯大學獲得生化遺傳學的博士學位，並獲得NASA的博士前獎學金。在博士學習期間，他還在美國梅因州巴哈伯的傑克遜實驗室研究哺乳動物遺傳學，參加了NIH獎項的學院培訓計劃；以及在百慕達聖喬治西的百慕達生物站參加海洋發育胚胎學課程的另一個學院培訓計劃，該計劃由NSF資助；還有在加州大學聖地亞哥分校的發育生物學分子技術研究所進行的第三個學院培訓計劃，同樣由NSF資助。

隨後，他在明尼蘇達大學作為行為遺傳學的博士後研究生，在那裡，他除了研究精神分裂症和阿茲海默症外，還通過協助編輯他的導師歐文·戈特斯曼（Irving Gottesman）博士的兩本書手稿學會了如何寫書（戈特斯曼博士在41年後回報了這個恩情，為這本《實用文本挖掘》寫了兩個教程）。在學術研究和教學職位之後，邁納博士在愛荷華大學進行了為期兩年的NIH博士後研究，專注於精神病流行病學和生物統計學，深入研究情感障礙和阿茲海默症。總的來說，他花了超過30年的時間研究和撰寫有關阿茲海默症遺傳學的論文和書籍（Miner, G.D., Richter, R, Blass, J.P., Valentine, J.L, 和 Winters-Miner, Linda. 《家族性阿茲海默症：分子遺傳學和臨床觀點》。Dekker: NYC, 1989；以及Miner, G.D., Winters-Miner, Linda, Blass, J.P., Richter, R, 和 Valentine, J.L. 《照顧阿茲海默症患者：家庭和醫療保健提供者指南》。Plenum Press Insight Books: NYC. 1989）。

多年來，他在包括堪薩斯大學、明尼蘇達大學、西北納撒尼大學、東納撒尼大學、南納撒尼大學、奧拉羅伯茨大學醫學院等多所大學擔任教授和系主任等職位。在1990年代，他還曾在加州帕薩迪納的富勒研究生心理學院和富勒神學院擔任老年心理學的臨床訪問教授。

1985年，他和妻子琳達·溫特斯·邁納博士（Dr. Linda Winters-Miner，這本書的幾個教程的作者）創立了家族性阿茲海默症研究基金會（The Familial Alzheimer's Disease Research Foundation，簡稱阿茲海默症基金會），該基金會成為組織本地和國際科學會議的主要力量，將來自幾個國家的阿茲海默症遺傳學領域的領導者聚集在一起，這導致了第一本有關阿茲海默症遺傳學的科學書籍的撰寫；這本書包括了來自1987年10月在俄克拉荷馬州塔爾薩舉行的第一屆國際阿茲海默症遺傳學研討會的100多位科學家的論文。在這段時間內，他還曾擔任位於俄克拉荷馬城的俄克拉荷馬醫學研究基金會的附屬研究科學家，與俄克拉荷馬大學醫學院合作。

邁納博士在正確的時機將全世界領先的阿茲海默症遺傳學科學家聚集在一起，當時哈佛大學、杜克大學、加州大學聖地亞哥分校、德國海德堡大學以及比利時、法國、英國和澳大利亞珀斯的多所大學開始發現他們認為與阿茲海默症相關的基因。

在1990年代，邁納博士被任命為俄克拉荷馬州州長的阿茲海默症工作小組，並擔任《老年精神病學與神經學期刊》（THE JOURNAL OF GERIATRIC PSYCHIATRY & NEUROLOGY）的阿茲海默症副編輯，至今仍在該期刊任職。到1995年，大多數這些顯性遺傳的阿茲海默症基因已被發現，而邁納博士自1980年代中期以來與華盛頓大學合作的基因是最後一個被識別的基因，位於人類基因組的第1號染色體上。當時，邁納博士在發現阿茲海默症的一些遺傳學目標後，決定做些不同的事情，尋找商業世界的一個領域，因為他已經分析數據超過30年，為StatSoft, Inc.擔任高級統計師和數據挖掘顧問似乎是一個完美的半退休職業。有趣的是（正如他的妻子所預測的），他發現商業世界比學術界有趣得多，在1999年舊金山的KDD數據挖掘會議上，他決定專注於數據挖掘。順便提一下，他在那裡首次遇見了鮑勃·尼斯比特（Bob Nisbet），鮑勃告訴他：“你必須認識這位聰明的年輕新星約翰·埃爾德（John Elder）！”幾分鐘內，鮑勃就找到了約翰並介紹了他，因為約翰也參加了這次會議。

隨著加里深入這個新的數據挖掘領域，並查看一般的統計學教科書，他看到了“實用統計書籍”的需求，開始撰寫章節並組織不同書籍的各種大綱。加里、鮑勃和約翰在KDD會議上不斷相遇，最終在2005年8月的西雅圖早餐會議上決定需要撰寫一本關於數據挖掘的書，並在那裡重新組織了加里的大綱，最終成為2009年由Elsevier出版的《統計分析與數據挖掘應用手冊》。然後，在2012年，他成為了Elsevier/Academic Press的第二本書《實用文本挖掘》的主編。2015年，這個系列的第三本書《實用預測分析與醫學決策系統》也隨之出版。這一切都要感謝加里的導師歐文·戈特斯曼博士，他在1970年時在加里與他一起進行博士後研究時播下了寫書的種子。

他最新的書籍於2018年發布，是2009年書籍《統計分析與數據挖掘應用手冊》的第二版（https://www.amazon.com/Handbook-Statistical-Analysis-Mining-Applications/dp/0124166326/）；以及2019年為普通人和決策者撰寫的書籍，標題為《醫療保健的病痛 - 預測治療 - 有效的解決方案！！！》，由Routledge / Taylor and Francis Group出版（https://www.amazon.com/HEALTHCAREs-OUT-SICK-PREDICTING-INNOVATIONS/dp/1138581097）。

邁納博士目前正在與斯科特·伯克博士合作撰寫第二本和第三本書，並不定期在加州大學爾灣分校教授預測分析和醫療分析課程。