How to do a thorough data investigation on any data set in Alteryx скачать в хорошем качестве

How to do a thorough data investigation on any data set in Alteryx

In this video I walk you through a pretty thorough data investigation step-by-step in Alteryx. I tell you why we do that and I show you how we go and build on our findings to go to the next. At the beginning of this state investigation we don't know whole lot about the data. But you will quickly see as I stepped through it that we can start to see patterns and we can start to draw insights on those patterns. Get a free stock like Apple, Exxon Mobil, GM, Ford, Tesla or Facebook worth up to $500 just by signing up and joining Robinhood using my link: https://join.robinhood.com/davidm16707 Then as we go from one data investigation tool to another in Alteryx, that these patterns build off one another and help us to draw some pretty interesting insights - for instance, the correlation between temperature and total sales or rentals in this data set. The data set I use in this video is the bike share data set from the University of California at Irvine and their data science department. Once you've completed your data investigation you now have the information you need to go and figure out your next step Weatherby to do a minimalized analysis or a much deeper dive. This will give you a lot of information and visuals along with charts and more to back up your findings. So, for instance, if you deem that the data is sufficient to do a much deeper dive with a custom process in R, you will have the data from this investigation to show multiple correlations and then be able to build off that and tell your users and boss about the value of doing a larger project on that data. What I always find interesting is that when I do a date investigation like this I can take data that I have not even looked at. That I have no idea there's any correlations or anything with it. All I have to do is input into Alteryx, maybe do a few little formulas or a select statement on it to change the data types for numeric fields and then run it through a date investigation like this and in minutes I have an analysis like you'll see here. Through this analysis were able to determine that there is a direct and large correlation between temperature and new, old (subscribers) and total sales (rentals). There also some other smaller correlations, but these are the main ones. In this video I walk you through the whole entire process step-by-step and you will see five different data investigation tools being used. We look at distribution plots, field analysis, heat maps, correlation analysis and more. One of my favorite parts is the heat maps because you can identify various segments or problem areas. In this analysis we discover that there's two main hot areas or zones in the heat map. The area in between these is a low area and an opportunity to do some additional marketing to bring that area up and address those lower rentals are sales. One you've done a thorough data investigation like we do here in Alteryx, you will be comfortable with results because they will just makes sense. For instance, not many people rent a bike for outdoor use when it's really cold out. The same holds true when the temperature rise to be extremely hot. As the heat grows the sale start to drop off again. I like to use a thorough data investigation like this to sell users, managers and department heads on trending and why this data set in question needs a deeper dive. Because it's great data and will be able to extract some really interesting learning from it. And a lot of time just learning that we do from the data is reusable in other future analyses. Another bonus is that this model I'm building right here is reusable for basically any data set. One of the great things with Alteryx is that it doesn't have the data limitations and row limitations that Excel does. I hope you found this video interesting and informational. Please like and subscribe see can be notified of all my other great videos I have coming out! Also be sure and check out my channel for all of my other great videos on everything from data analysis, data science, forecasting, dat wrangling and more! Thanks again and God bless!