home | library | feedback

Uploading data

Before we can start the modeling we need some data. We welcome you to use your own data. However, getting your data read correctly into the systems is one of the most difficult parts of the data analysis tools. While we hope that it is the most difficult part of this "tool" too, we have tried to give clear instructions of the data format accepted.

Sending the data

You must use the form below to send your data to us so that we can analyze it. If you are taking your first trial on classification modeling, please read all the instructions before submitting your data.

The easiest and safest way is to first click the Browse-button below. That pops the file selector with which you can locate your data file. After closing the file selector, please check that the text field below contains the correct path to your data file. When so, press the "Send data"-button and we will tell you if everything went OK.

You may want to fill the optional data set name field so we can use the data set's name in appropriate places.

Data upload form

Data set name:

Data format

The format of the data this system is willing to accept resembles SPSS free format. It means simple text file, in which the first row contains names of the variables and the subsequent rows contain the data one case in each row. The names of the variables in the first row and the fields in data rows should be separated by tabulators. For the details about the accepted format (how to code missing data etc.) take a look at the format specification.

Data file size and other minor issues

To provide reasonable service to all the B-Course users we have set the maximum size of data file to be one megabyte. Even within these limits the large data sets take longer to upload and analyze than small ones, so if you have a large data set some steps may take a while to perform. Technically there is no lower or upper limit for the amount of data (not for number of cases or number of variables). However, that does not mean that the results or the analysis are independent of the size of the data.


  B-Course, version 2.0.0
CoSCo 2002