Data Analysis best practices by Sherlock Holmes! - My Datafication

26 March, 2017

Data Analysis best practices by Sherlock Holmes!


Did you miss me? Did you miss me? Did you miss me? If you did not get the reference from the Sherlock series, check below for the episode reference (spoiler alert). As you have probably imagined, I have recently watched the Sherlock series and as a big Sherlock Holmes fan I have enjoyed the stories of the genius detective and his loyal friend Dr. Watson. What I did not realize until now though, was how many things Sherlock teaches us about data analysis. I have taken my notes about Sherlock's best practices and would like to share some of them with you.

Lesson #1: Collect the data that ARE useful

Many cases solved by Sherlock required knowledge from different scientific areas. For example, he can distinguish 140 types of cigarettes, smoking and cigars, which proves to be something extremely useful to know in some cases. Similarly, every data analyst (and user in general) can use internet as the source of any useful information. However, there are available more data and knowledge that may be useful in a case, i.e. analysis project. Thus, it is important to distinguish important and useful knowledge from useless information. Sherlock did not know that the Earth resolves around the sun, because it is not important for criminal cases. So, collect only the data that you need and remove trash data! 

Lesson #2: Deduction is reasoning backwards.

This is actually a phrase from the story, repeated many times by Sherlock in order to teach Dr. Watson about the science of deduction. In all cases, he collects the data, thinks synthetically and connects the sequence of events revealing a story with all pieces connected and justified. However, it might be easier for people to avoid situations that are difficult to explain or rely on absolute truths i.e. conventional wisdom that has never been proven to us. On the other hand, deduction does not allow such shortcuts. You have to follow the facts so that you reach correct conclusions, even if they surprise you. "When you have eliminated all which is impossible, then whatever remains, however improbable, must be the truth" (Sherlock Holmes). So, in all projects never guess! Deduct!

Lesson #3: Use FACTS to stimulate knowledge.

Similar to the Deduction lesson, Sherlock emphasizes the need to collect facts and data to reach conclusions. "It is a capital mistake to theorize before one has the facts. Insensibly, one begins to twist facts to suit theories, rather than theories to suit facts" (Sherlock Holmes). Unfortunately, there are times when people try to justify a theory using the data instead of using the data to make theories. This problem may arise  in companies if managers need justifications for their decisions, or researches if analysts try to validate theories or any other similar cases. It should be part of analysts' ethics to never use data to prove theories, but always analyze them unaffected by existing thought or theories. Data should always lead us! 

Lesson #4: Concentrate

Having collected the data and the facts, Sherlock plunges into his own world. He isolates himself in order to concentrate on the facts and build his reasoning. In our modern world, with so many distractions, it might be difficult to isolate ourselves in order to concentrate on a given analysis. However, it is important to train this skill to eliminate influences and diversion to optimize our work. Focusing 100% on one task may help identifying all useful facts and leading to the correct conclusions.
I hope you liked the article, elementary, my dear Watson! Sherlock Holmes is one of the best data analysts (or maybe the best, given that he analyzes everything using only his brain). And for sure, we could use his lessons to improve our methods and way of working. If you have read the Sherlock stories and have noticed another interesting lesson(s), please share it with me in the comment section below!

1. Sherlock Season 3, episode 3, from 1:27:24. Jim Moriarty's video where he asks 'Did you miss me?' again and again!

If you liked the post, don't forget to like "My Datafication" on Facebook to stay up-to-date with the latest posts! Last but not least, if you would like me to elaborate more about any of the above mentioned topics, or if you have something interesting in mind, leave your comment below or in my Facebook Page.

20 comments:

  1. I guess you got your hands on IBM Watson??
    Katerina, I'm a fellow student of MSBA (AUEB) on my last year of studies.

    All my best wishes to your goal!
    U r an example that keeps my hopes for a data career abroad alive!

    ReplyDelete
    Replies
    1. IBM Watson! Been there, done that! But how did I forget about it while writing the post? :) Thank you very much for your comment! I wish you success, too!

      Delete
  2. Hello Katerina Kalaidopoulou,
    The Article on Data Analysis best practices by Sherlock Holmes is nice .It give detail information about it.Thanks for Sharing the information about it. data science consulting

    ReplyDelete
  3. likewise need them to improve their business utilizing information. Organizations are paying significant compensations for gifted experts and since the speed of vocation development is huge, there's no better time to learn Data Science NOW!ExcelR Data Science Courses

    ReplyDelete
  4. Great post i must say and thanks for the information. Education is definitely a sticky subject. However, is still among the leading topics of our time. I appreciate your post and look forward to more.

    Correlation vs Covariance

    ReplyDelete
  5. As consumers, people are weary of purchasing from online companies, especially in the case where other review sites can hinder their trust in the company. Safer Reviews

    ReplyDelete
  6. I am looking for and I love to post a comment that "The content of your post is awesome" Great work!

    Simple Linear Regression

    Correlation vs covariance

    KNN Algorithm

    Logistic Regression explained

    ReplyDelete
  7. Really nice and interesting post. I was looking for this kind of information and enjoyed reading this one. Keep posting. Thanks for sharing.

    data science interview questions

    ReplyDelete
  8. This is my first visit to your web journal! We are a group of volunteers and new activities in the same specialty. Website gave us helpful data to work. site de rencontre sérieux

    ReplyDelete
  9. "It is amazing and wonderful to visit your site.Thanks for sharing this information,this is useful to me...
    data science courses"

    ReplyDelete
  10. It feels entirely appropriate. Every one of smaller aspects were created through many record education. I really like the application plenty Fateh soft

    ReplyDelete
  11. Thank you for taking the time to publish this information very useful! Outlook CRM

    ReplyDelete
  12. I’m excited to uncover this page. I need to to thank you for ones time for this particularly fantastic read!! I definitely really liked every part of it and i also have you saved to fav to look at new information in your site. data is sacred | data trends

    ReplyDelete
  13. data scientist course in pune with placement was never so easy and adaptable to everyone but here at Excelr We teach you numerous ways of doing Data Science Courses, which are way easy and interesting. Our experienced and expert faculty will help you reach your goal. 100% result oriented strategies are being performed; we offer Data Science Course in pune


    data scientist course in pune with placement
    Data scientist certification
    Data scientist courses

    ReplyDelete
  14. Phone or Online Consultation with respect to what steps ( Precautions and information misfortune systems ) to take when Data Crisis Happen. We will disclose our recuperation interaction to guarantee the media is maneuvered carefully after an accident.Manchester data recovery

    ReplyDelete
  15. There are many reasons why it's important to verify Social Security number information on your employees such as to make sure that you are hiring legal workers, ensuring accurate wage reports, and ensuring that your employees' wages are properly credited to their SSA earnings records. A simple typo could have disastrous long-term effects on an employee. numbersdata.com

    ReplyDelete
  16. Data Recovery begins upon once its affirmed from Client. Data Recovery Melbourne

    ReplyDelete
  17. Think of it as a home video that you have decided you don't want anymore. Until you actually tape over it, it's still there. hard drive recovery melbourne

    ReplyDelete