Customer Behavior Prediction Using Web Usage Mining
Web usage mining involves first recording behavior and flow of customers on a website and then mining through this data for behavioral patterns. It is an important part of ecommerce world that allows websites to go through previously recorded web traffic data. Ecommerce sites analyze this data in order to provide better performance and also suggest better products and services to customers by identifying them next time. The system is tuned to record web shopping/buying patterns and track various analytics data that tend to provide future prediction statistics. The system scans for user budget tracking, tallying to previous years, user bounce rates- number of users returning from payment page and other site usage factors. Factors like returning users allow site owners to make appropriate changes and give the customer exactly what is needed. This allows for more customer acquisition and thus more profitability. Ecommerce sites need to survey and mine for previously recorded data to check their website performance and constantly optimize it as per customer needs.
Reference:
- http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6149134&ranges%3D2011_2014_p_Publication_Year%26queryText%3Decommerce+behaviour
- http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6703147&ranges%3D2011_2014_p_Publication_Year%26queryText%3Decommerce+behaviour
Software Requirements:
- Windows Xp, Windows 7(ultimate, enterprise)
- Sql 2005
- Visual studio 2010
Hardware Components:
- Processor – i3
- Hard Disk – 5 GB
- Memory – 1GB RAM
Advantages:
- The system is easy to install.
- It will help the proprietor to assume next step of user.
- He can develop its future plans based on this study.
Disadvantages:
- It uses lot of memory on the server to store the data.
- It is not so accurate.