A Web Traffic Data Analysis System based on Statistical-Analytical Report Generation

To increase the business value in terms of quality of service, it is quite necessary to have aspects of the performance of the system. The website ownership always demands to perform certain analysis so that they can get maximum profit through their business. However, conducting meaningful data analysis requires a proper understanding of the data’s narrative and/or interpretation as numbers alone do not make any sense. Improper understanding of data could result in wrong output. To overcome this, specialized knowledge and expertise is the possible solution. Therefore, the authors created an automated system for analyzing web traffic data from mobile devices and web. The descriptive statistical-analytical techniques had been utilized in the report-generation process, but now the system employs a technique called Bootstrap Aggregator to automatically identify customer needs and generate data analysis reports. Furthermore, the researchers compared the output reports generated by two different algorithms: Deterministic Linear Decision Tree (DLDT)-for analytical and Random Forest-for automatic. Their findings have revealed that both algorithms produce reports with a high degree of user satisfaction and acceptance.