For example, I want it to help me process the data table following data. I used to work on in 2009, which contains the performance of each ad space at that time. All I have to do is upload this Excel file to it, and it will pre-process the data table.
Of course, you will find that the above clustering is a bit unscientific. In fact, it is not the fault of Julius, but the data I provided has some problems. For example, some Conversion Rate data is greater than 1, and even reaches 2.0. Some CPMs also reach several tens of percent. So I asked it to exclude these abnormal data and help me cluster again.
So I got the new clustering result below. I took a closer look and it was really good. The data was accurate and the clusters were clear and reasonable. I no longer had to ask my data scientist colleagues to help me process the data. I had this tool and could do it myself!
Then upload it again data table
Of course, it can’t just do some data processing. I wanted how to choose the right developer to get the best mobile app development rates to test its autonomous analysis capabilities. So I typed the following paragraph to it:
That is, tell me which website performs best among the data?
Guess what? It said that my data has been uploaded for more the beginnings of modern treatment plants than 1 hour and has been deleted for privacy and security reasons!
After uploading, it gave a conclusion, but this conclusion was obviously not smart enough. It was still arguing about the “Actual Cost” that I just asked it to draw, so I was not satisfied with the result.
So I asked again
“You can’t just look at the actual cost, you of germany cell number the have to look at other data as well.”
So it really looked at the other data and came up with some simple conclusions.