Application development for improving website usability by web mining methods için kapak resmi
Application development for improving website usability by web mining methods
Başlık:
Application development for improving website usability by web mining methods
Yazar:
Türker, Merve Ceren, author.
Fiziksel Tanımlama:
xi, 78 leaves: illustrarions, charts;+ 1 computer laser optical disc.
Özet:
The explosive growth in website traffics and website usage data has resulted in the amount of valuable information contained to have a similar uptrend in web usage logs. With the increasing competition between websites, mining web usage logs to discover meaningful information is needed more than ever. Web usage mining is the procedure of using data mining methods to discover insightful patterns in web usage logs. The discovered information helps understand how users behave on the website and their needs. One of the most popular algorithmic approaches of pattern mining on web usage data is the Fp-growth algorithm. For larger volumes of data, the algorithm is generally applied to execute in parallel. Measuring and comparing performances of applications is difficult, because the algorithm performs different on usage logs with different characteristics. The characteristics of usage logs are highly related with the type of the website. In this paper, we have investigated how different characteristics of web usage logs effect the performance of the parallel Fp-growth algorithm. Five datasets with varying log characteristics were used in order to represent different business models. The results suggest that the performance is highly correlated with the number of items, number of frequent items, transaction length, similarity between frequent patterns, minimum support value and size of the log file.
Yazar Ek Girişi:
Tek Biçim Eser Adı:
Thesis (Master)--İzmir Institute of Technology: Computer Engineering.

İzmir Institute of Technology: Computer Engineering--Thesis (Master).
Elektronik Erişim:
Access to Electronic Versiyon.
Ayırtma: Copies: