Microsoft Business Intelligence. Is this what I am trying to make possible?
I am tasked with analyzing the logs table of my website. This table contains the user's click path across the entire website for a given session. My company wants to understand / identify trends based on our users' clicks. In doing so, identify user groups that use a specific "click path" based on age / geography, etc.
As you can tell from the title, I am completely new to BI and its capabilities, so I was wondering:
- Have our goals been achieved?
- How should I do it?
I am currently reading books online as well as other e-books I have found. All indications seem to suggest that this is possible through sequence clustering. Even though I'm currently fine with the exact implementation and settings. So if anyone has first-hand experience with such an endeavor, I would be awesome if you could share it here.
Hooray!
source to share
What you are looking for is called Rule Mining Association . I am not particularly familiar with BI, but I suggest you take a look at Weka , which contains several implementations of the Apriori algorithm and its variants.
source to share
It won't help you with your existing log files ... (but it's an alternative if your search for an answer fails)
Google Analytics is free and you can set up a few custom variables {age, etc} and see where the traffic is going. (you won't be able to see what an individual user is doing.) Not exactly exactly when you are trying to do, but free and can be done to be close to what ur is looking for
If you want really good analytics take a look at Omniture (expensive) but its top tier for creating complex website reports. It is used in many e-commerce scenarios tracking how a user logs in and interacts with a site + much more ~
There are a lot of website analytics out there, before you "roll" yourself, take a look at some of them, they can help you focus on your own goals.
source to share
It seems you can use neural networks for this task. Possibly persetrons .
I have experience with neural networks, but I'm not an expert.
I highly recommend the book Collective Intelligence Programming: Building Smart Web 2.0 Applications. Check it out even if you don't know Python.
source to share
Start with open source or commercial web analytics software first (google for that) as reading web server log files is non-trivial
Some allow you to map data to other tables (your users table with age, etc.) or mix your own solution to map web session logs to other data.
Except that regular SQL queries will solve your problem with analysts for example.
select user.id
from user, log l1, log l2, log l3
where user.id = l1.userid and l1.type = first step
and user.id = l2.userid and l2.type = next step
and user.id = l3.userid and l3.type = last step
and l1.sessionid = l2.sessionid and l2.sessionid = l3.sessionid
Loading raw data into a BI structure might not make it much easier. Loading the results of queries like this into a BI framework would make scense
Depending on your web application, you may have trouble identifying actaul sessions if they have a long session ID, etc. or changing the session id. If this is a problem, you need to translate the web analytics into actual web server code so that you can simulate long term state and recording, and instead
source to share