First, Did David Bowie hide the secret to successful model deployment in the lyrics to his worldwide hit “Space Oddity”? He very well may have. CRISP-DM is comprised of six stages, the last of which is model deployment. This is common knowledge to many data scientists that work with CRISP-DM in mind every day, and yet even experienced datanauts may benefit from an exploration of this process.
By the way if you haven’t heard this remix of the song it goes well in the background while you read the article… give it a try
What exactly does it mean…
Agility has become a buzzword in the data science and analytics community as the demand for flexible storage solutions and the increased data retrieval speed continue to climb in tandem. Historically speaking, when these types of discoveries are made, it has induced dramatic and, at times, traumatic change. Something as simple as discovering a valuable piece of metal, a new technology, or a novel tool has been the impetus of mass migrations, gold rushes, industrial revolutions, and wars. In the twenty-first century, agility is vital.
Though the streets remain relatively calm, the skies are void of black factory…
When it comes to defending your most valuable assets, there is no such thing as going too far. People may argue that there is a line where the expense incurred has no measurable ROI, and to a certain extent, this is true. Everything is limited by capital investment; however, there are established methods of minimizing cost and maximizing results. Figure 1 shows a preview of what this post will cover.
Figure 1. Cirrus Word Cloud of This Discussion Post (Waters, 2020).
When I worked as a Realtor,
I would often find myself charting out my cold calls and keeping needless statistics that, in retrospect, held little or no future value. This seemed like a
This is an essential question for companies to ask in the year 2020. What reasons exist today that have motivated the modern marketers’ push to tap into this online town-square, and do the challenges encountered along the way make the opportunities worth the risk? The answer may not be as simple as a list of tips, and there is one preliminary question that needs to be answered: what is marketing without social media? If your company’s marketing department is excluding social media from your marketing strategy in our current moment, then your company will fall behind competitors within the next…
In my previous tutorial, we discussed how to use Hazel for basic automation and file sorting. This article will dive deeper into several rules using Hazel as your personal assistant that saves you space and salvages your external hard drives.
First, create a testing folder on your drive called “HAZEL TESTING AREA.” This is not a suggestion. I don’t want you to lose all your precious files because of a simple syntax error (yes, that can happen). …
The third wave of COVID-19 is coming… or so we hear. But the most deadly enemy of public health, according to one doctor I spoke with yesterday, is the lack of ICU beds. With patients not being seen by medical staff for common, but sometimes deadly, illnesses, the death rates are sure to go up. This is almost like a second-hand COVID-19 fatality. According to this doctor, it is paramount, now more than ever, to maintain high standards of health for ourselves to stay out of the hospital. So how can we stay ahead of the curve when all data…
# We need to flip the rank for interceptions
1 - data_radar_filtered['Int_Rank']
This will result in a syntax error if copy/pasted into your code editor. Make sure to remove the new line after the "="
SyntaxError: invalid syntax (<ipython-input-65-480e5b366298>, line 3)
File "<ipython-input-65-480e5b366298>", line 3
data_radar_filtered[i + '_Rank'] =
SyntaxError: invalid syntax
for i in categories:
data_radar_filtered[i + '_Rank'] = data_radar_filtered[i].rank(pct=True)# We need to flip the rank for interceptions
data_radar_filtered['Int_Rank'] = 1 - data_radar_filtered['Int_Rank']
Suppose we had used predictive algorithms to scan legacy medical data and patterns of historical epidemic outbreak frequency back in 2019. In that circumstance, our country might have been able to preempt the recent global pandemic with preventative measures and saved thousands of lives.
This statement may seem grandiose and overly ambitious to some. Yet I feel confident that others may actually like, if not agree with it. …
I would like to tell you the story of two technological entities that fell in love. It happened like many good love stories do, a long time ago; back when analytics was a nascent field, and much of the data was unstructured and analog. The text and free-form data were largely unable to be captured and were thus unknowable in a digital sense, rendering only marginal insights. Now, with the marriage of faster computers, better OCR scanning technologies, more intelligent artificial intelligence algorithms, and higher demand for improvement at the organizational level text analytics, and text mining are finally joining…