I am new to datamining and RM, I want to relate two data tables. Each table has a unique key of date and number. Thus the tables should be related via this key. How can this be done in RM ? Thanks Arthur
Hi Arthur, what exactly do you mean by relate? I guess it's the same you do, when you join some relational data base tables? We have an operator called "Join" which will do this.
Thanks Sebastian, The problem is before the join is executed. Table A: has as field names: Flight Date Delay Pax Table B: has as field names: Flight Date Report In the result of the Join I want to have a table C which has as rows the flights that appears in both tables and as fieldnames the fieldnames that appear in both tables. So I can see how many pax had delay. Table A and Table B have no Id. They are linked by Flight and Date. How is this done in Rapid Miner, Thanks Arthur
Hi Arthur, so neither flight and date are unique, but flight + date is unique? Then I would suggest building a new id column from the values of flight and date column. This column can be used for the join. Let make a short sample process:
After we generated some data and transformed it for having an comparable situation, we build a new id like attribute by concatenating the values of this two attribute:
Thanks for your process. It is not yet the solution but we are getting closer. I can never get the same Id in both file for the same FlightNr and Data combination because the number of records is different in both files. I try to explain the problem in different words:
File A has fields: FlightNr, Date, Pax, Fuel File B has fields: FlightNr, Date, Event File B has less records. Only the flight that have an event are in file B. I want to get File C with fields: FlightNr, Date, Pax, Fuel, Event On most flights no events are recorded so many event fields will be empty in file C. Resulting file example: Flt112,1-1-2010,34,2300,? Flt116,1-1,2010,45,2900,turbulence Flt112,2-1-2010,23,2100,? Flt116,2-1-2010,41,2800,?
Answers
what exactly do you mean by relate? I guess it's the same you do, when you join some relational data base tables? We have an operator called "Join" which will do this.
Greetings,
Sebastian
The problem is before the join is executed.
Table A: has as field names: Flight Date Delay Pax
Table B: has as field names: Flight Date Report
In the result of the Join I want to have a table C which has as rows the flights that appears in both tables and as fieldnames the fieldnames that appear in both tables. So I can see how many pax had delay.
Table A and Table B have no Id. They are linked by Flight and Date.
How is this done in Rapid Miner,
Thanks
Arthur
so neither flight and date are unique, but flight + date is unique? Then I would suggest building a new id column from the values of flight and date column. This column can be used for the join. Let make a short sample process:
After we generated some data and transformed it for having an comparable situation, we build a new id like attribute by concatenating the values of this two attribute: I hope, it will help you.
Greetings,
Sebastian
I can never get the same Id in both file for the same FlightNr and Data combination because the number of records is different in both files.
I try to explain the problem in different words:
File A has fields: FlightNr, Date, Pax, Fuel
File B has fields: FlightNr, Date, Event
File B has less records. Only the flight that have an event are in file B.
I want to get
File C with fields: FlightNr, Date, Pax, Fuel, Event
On most flights no events are recorded so many event fields will be empty in file C.
Resulting file example:
Flt112,1-1-2010,34,2300,?
Flt116,1-1,2010,45,2900,turbulence
Flt112,2-1-2010,23,2100,?
Flt116,2-1-2010,41,2800,?
Thanks
Arthur