"Web Mining Operators"

fbarth

所有的人,

I'm trying to use the Reader Server Log Operator, but I cannot find any example about the config file (a necessary parameter of Reader Server Log Operador).

Anyone can tell me where I can find an example? I searched intohttp://polliwog.sourceforge.net/, but I couldn't find.

Best regards,

Fabrício J. Barth

land

Hi,
perhaps this one will help you. Detailed instructions are available on the url you already posted.

openQuote="["
closeQuote="]">
value="en/US" />
value="dd/MMM/yyyy:HH:mm:ss Z" />



openQuote='"'
closeQuote='"'
escapedBy="\" />








openQuote='"'
closeQuote='"' />


openQuote='"'
closeQuote='"'
escapedBy="\">
value="user-agent" />

makchishing

fbarth wrote:
所有的人, I'm trying to use the Reader Server Log Operator, but I cannot find any example about the config file (a necessary parameter of Reader Server Log Operador). Anyone can tell me where I can find an example? I searched intohttp://polliwog.sourceforge.net/, but I couldn't find. Best regards, Fabrício J. Barth

I have server logs that are zipped to gz file(320 MB only),
If upzip to text file, around 3GB. > <

Can rapidminer support read server log for a zipped format?

land

Hi,
实际RapidMiner无法读取压缩日志文件。的course it would be possible without much work, but what benefit would result from that? Before handling the data RapidMiner would have to extract it. So the data would be extracted not just once, but each time you process it...
If you need to process the data in an online fashion and extract them each time the process is executed to work on the most recent data, just use the execute operator for shell commands.

Greetings,
Sebastian

makchishing

Sebastian Land wrote:

Hi, 实际RapidMiner无法读取压缩日志文件。的course it would be possible without much work, but what benefit would result from that? Before handling the data RapidMiner would have to extract it. So the data would be extracted not just once, but each time you process it... If you need to process the data in an online fashion and extract them each time the process is executed to work on the most recent data, just use the execute operator for shell commands. Greetings, Sebastian

Thanks Sebastian Land ,

It is very beneficial,
Save harddisk space, save the time for extraction, lots of time is wasted to wait for the extraction.
It is very easy to read zipped file by java, as I searched from web.
>< I have found a simple program to solve that.
http://www.java2s.com/Code/Java/File-Input-Output/Readsomedatafromagzipfile.htm

land

Hi,
I'm completely aware that it is pretty easy to read zipped files. I only doubt that it is useful, since you will have to extract the data anyway. What does it make for a difference if you extract it once before reading the data or during reading the data? If you execute the process twice, you will have to do the extraction twice. So where does the benefit comes from?

Greetings,
Sebastian