How to proceed with Data Mining assingment

[BruLee]

Hey guys,

I need help with a data mining assignment. I have attached dummy data that will help you understand how my data is structured.

What I need to achieve:

1. Being able to filter the data by multiple queries (people who have used the app "MO" + they are from "USA")
2. Being able to filter the data by interests as well (people who have an interest in "Movies" & they are from "GB")
3. After the filtering is done, I need to know the numbers of records that meet my criteria as well as their interests in the case of 1. So I want to know the top 5 interests of the users who used the app "MO" & that are from "USA".
   After that, creating charts to show these numbers would be of added value.

My objective from this assignment is to answer the question HOW MANY & not WHO. So if I query "MO" users + from "GB" then I want to know how many people that fit these criteria exist as well as what are their top 5 interests.

Which direction should I take to achieve this objective? Should I use pivot tables? (I'm having problems with the interests data being divided in 3) Should I not use excel at all for this?

Awaiting your answers.

Cheers.

data-mining.xlsx

[BruLee]

Bump please

[Andy Pope]

If you want to use pivot tables in excel then you would do weel to change the layout of the data.
Instead of 3 fields for interest expand the records with a single Interest field.

See sheet2 for new layout and sheet5 for PT

Whilst doing your analysis you may need to move the fields around between row/column/page fields to get the information required.

[BruLee]

Thank you for this, looks good. How would I suggest that I expand the records if I have a lot of rows? Should this be acheived with Macros?

[Andy Pope]

either manually or code or formula whatever is easiest

[romperstomper]

Cross-post here.