Abstract
The article describes the algorithms for building recommender systems based on the analysis of data on grocery purchases of users of one of the largest online retailers. Using modern methods of data storage and analysis, effective recommender systems allow, in particular, to form the customers' interest and to increase the value of the average bill in individual orders. The article describes a built-in hardware and software system on Amazon cloud web services. Using this system, the initial data set was studied, typical examples of recommendations were constructed and the ALS and SVD algorithms were compared. We considered the use of frameworks Apache Hadoop and Apache Spark for the analysis of large format data on consumer baskets. The article analyzes the peak days of the week and workload during the day. Found popular product categories. We studied the demand for various groups of goods by day of the week and the frequency of purchases. The dependencies between the first and subsequent orders, popular products for the first and subsequent orders, and also changes in customer preferences over time were identified.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
