Hello,
I am trying to make some data mining projects, but I am stuck.
I read books and other literature about data mining. It offers a lot of theoretical information that I've become familiar with. But, all examples are quite similar and very simple.
Now when I have to make practical application I am quite confused and not sure what to do.
I would be very thankful if anyone can give me some hints about my problem or information where can I found some useful examples.
I’m working with DB Intelligent Miner. Tool I am working with is not important, just it offers many data mining techniques and variety preprocessing tools.
Databases I am working with are, if I am not wrong, sample DB2 databases, such as GOMR, GOSL.
GOMR consists of tables with data about retailers, their demographic data, semi annual reports, etc.
One of the things I have to do is to analyze data in order to find quality of sales depending on retailer age and segment.
Reports for retailers are for half of year, and include data about incomes and outcomes (fixed assets, current liabilities, net income, expenses, etc.).
Rating of success should include income and outcome. How should this influence my results? I am not sure should I calculate an additional field that includes all amounts (additions and subtractions). I was thinking about weighting columns with +1 and -1. But then I realized it's, probably, wrong way to denote that some column reduces "success" (has negative impact).
Is it good idea to weight columns and to give lower weight to (for instance) long term debt, because I have to rate current success of retailers?
GOSL includes information about orders, returns, sales target.
What algorithm should I use for finding does returning of product affects sales of it?
Thanks in advance.
I am trying to make some data mining projects, but I am stuck.
I read books and other literature about data mining. It offers a lot of theoretical information that I've become familiar with. But, all examples are quite similar and very simple.
Now when I have to make practical application I am quite confused and not sure what to do.
I would be very thankful if anyone can give me some hints about my problem or information where can I found some useful examples.
I’m working with DB Intelligent Miner. Tool I am working with is not important, just it offers many data mining techniques and variety preprocessing tools.
Databases I am working with are, if I am not wrong, sample DB2 databases, such as GOMR, GOSL.
GOMR consists of tables with data about retailers, their demographic data, semi annual reports, etc.
One of the things I have to do is to analyze data in order to find quality of sales depending on retailer age and segment.
Reports for retailers are for half of year, and include data about incomes and outcomes (fixed assets, current liabilities, net income, expenses, etc.).
Rating of success should include income and outcome. How should this influence my results? I am not sure should I calculate an additional field that includes all amounts (additions and subtractions). I was thinking about weighting columns with +1 and -1. But then I realized it's, probably, wrong way to denote that some column reduces "success" (has negative impact).
Is it good idea to weight columns and to give lower weight to (for instance) long term debt, because I have to rate current success of retailers?
GOSL includes information about orders, returns, sales target.
What algorithm should I use for finding does returning of product affects sales of it?
Thanks in advance.