Familiar. Collaborative. Managed.
We’ve all heard about “Big Data” by now but what does it really mean? Big Data is the term used to describe the process of applying serious computing power and the latest in machine learning and artificial intelligence, to massive and often highly complex sets of information.
In a recent example, the Microsoft Data Explorer team used food establishment inspection data to create a wonderful example of making “Big Data” easy to surf. For those of you in the Seattle area, this is an interesting application to use before picking a Father’s Day restaurant. Another more serious example is discussed by the same team and shows an interesting crime rate application that would benefit business owners and potential home buyers.
To help you better understand the impact of big data, Microsoft has recorded an exclusive webcast briefing, "Tools to Turn Emergencies into Knowledge,” produced in association with CIO Magazine that you can view now.
For more information see the moderated discussion and webcast in the article “Turning 911 to 411”.
I am a novice of dataming. Now I have a problem is to predicte: Who will buy something in the next month and which brands they will buy ?. The data is from a online shopping website, just like eBay. some data is given as blow(just a sample here) :
user_id brand_id type visit_datetime
10944750 13451 0 2013/4/15
10944750 21110 0 2013/4/17
23221235 21134 2 2013/4/12
In fact, There are a lot users and brands. The records are about 20 thousands, datetime from 2013/4/15 – 2013/8/15.
the “type” has means four actions : 0 is click, 1 is purchase, 2 is Favorites , 3 push to shopping cart.
So, we are going to predict type=1, and the time is September 2013. The output likes:
user1 -> brand1,brand2
user2 -> brand2.
I can only use sqlserver and excel do some basic analyze. I just want to know The right way to analyze these data and Which Algorithm or model to use. Thank you very much!
You can see this question on stackflow also:stackoverflow.com/.../the-data-ming-on-users-online-shopping-records