Analysis of Click Usage Data

Daly, Joanne (2014) Analysis of Click Usage Data. Diploma thesis, Dublin, National College of Ireland.

PDF (Diploma)
Download (2MB) | Preview


This analysis project is being done in conjunction with The TAS Group who have a real world large data-mining problem that needs to be addressed. The TAS Group provides sales automation software to companies through its product Dealmaker®. They collect customer usage data from Dealmaker® customers and store it in a usage database.

Tracking customer usage is essential to understanding usage patterns. High usage has proven to be crucial in driving up adoption, which ultimately leads to an increased chance of licence renewals. Currently the usage data is consumed through text-based reports, which only provide information of the total clicks for a user and high level summaries.

The primary objective of this project was to devise a method to analyse patterns in Dealmaker® customer usage data. A number of data mining techniques were considered for the analysis such as Clustering and Association Rules.

During the analysis phase these techniques along with a Pareto Analysis were utilised in order to determine if significant usage patterns exist.

The high level summary analysis provided rudimentary insight which quickly identified those companies that regularly use the product from those that don’t. This information enables the TAS Group to targeted low usage customers and take steps to remedy the situation.

Detailed analysis through association rules provided interesting insights into the usage patterns of high usage customers. The analysis identified a low concentration of high usage customers that appear to follow a consistent pattern of usage. This suggests a localised but consistent usage pattern which is illustrated by the low support levels and high confidence levels found in the association rules. This potentially provides a mechanism to measure usage performance in a company or group of users by evaluating their support level and ensuring that the confidence levels remain high. Any drop in confidence or support levels may indicate a deviation from prescribed usage patterns.

This essentially provides The TAS Group with a method of analysing usage patterns. It also enables measurement of change in usage behaviour, which was a primary objective of the project.

Item Type: Thesis (Diploma)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Divisions: School of Computing > Higher Diploma in Science in Data Analytics
Date Deposited: 15 Dec 2014 11:55
Last Modified: 15 Dec 2014 13:06

Actions (login required)

View Item View Item