How "can you use these products" in online stores?

Question

Some e-commerce platforms have an offer function where they inform you as soon as you have an item in the basket that “you might like this product as well.” Some, like Amazon, rely on pre-existing customer behavior data, and have a feature called “Customers who bought this product also purchased,” but some seem to offer it in other ways.

What are these “other means”? What algorithms do they use in online stores for this opportunity?

Answer 1

Most offers on e-commerce pages are created using some kind of recommendation system ( http://en.wikipedia.org/wiki/Recommender_system ). There are tools like Mahout ( http://mahout.apache.org/ ) that already have implementations of the most common approaches.

Answer 2

They use data mining, and this particular algorithm you are asking for is called the “nearest neighbor” algorithm.

Here is a link to an article I wrote about the algorithm (like so many others).

http://www.ibm.com/developerworks/opensource/library/os-weka3/index.html

Answer 3

The process is called Business Intelligence , the data will be stored in the data warehouse, and the business intelligence process can be used using a product such as SSAS . This process will include grouping data volumes (Who bought what and when) into data cubes . The analysis is performed on these cubes and is used to compare your purchases with others who bought the same product, then he will recommend his purchases (the other customers who bought it also bought this product .... Item X). Other various AI algorithms are used to compare models according to other customer trends, for example, how they shop, where they click, etc. All this data is accumulated and then added to the data cube for analysis.

The data mining algorithms are described below; you can find the Decision Tree Modelling algorithm , which defines BI trends and patterns (in this case, Recommendations):

http://msdn.microsoft.com/en-us/library/ms175595.aspx

Answer 4

best book about this kind of algorithms: collective intelligence programming

Answer 5

As some of the previous people have said, this is called a recommendation mechanism. It is also referred to as a collaborative filtering method. There are few tools that do this; Mahut is one of them. Check out the blog that I wrote about the use case where we use Mahout and Hadoop to create a recommendation mechanism . As a precursor to this, I also wrote Component architecture on how each of them is suitable for a data retrieval problem.

Mahout will work offline as well as with Hadoop. The decision to use one really comes down to the size of the historical data that needs to be extracted. If the data size is of the order of Terabytes and Petabytes, you usually use Mahout with Hadoop. Weka is another similar open source project. All of them fall into the category called machine learning systems. Hope this helps.