Basic information about a final thesis
|Type of thesis:|
Modelling Changes in User Short term Behaviour on a Website
|Written by (author):||Ing. Ondrej Kaššák, PhD.|
|Department:||Institute of Informatics, Information Systems and Software Engineering (FIIT)|
Final thesis progress:
|Final thesis was successfully defended.|
Additional informationAdditional information about the final thesis follows. Click on the language link to display the information in the desired language.
Language of final thesis:
|Title of the thesis:||Modelling Changes in User Short term Behaviour on a Website|
User modelling within a website represents a process of capturing chosen characteristics based on observation of user behaviour, his personal traits or preferences. It is also a process of transformation user behaviour (on the level of actions) into aggregated form revealing complex relations, which are not visible directly from captured actions. Its aim is to create an abstraction of real user behaviour - a user model. The user models are typically used as an input for chosen task enabling personalization, prediction, recommendation or in general an improvement of interaction quality between the user and the website. Nowadays, the Web represents an information space, in which the content is dynamically changed. In several domains such as news, multimedia or e-commerce, the items arise quickly and remain actual only for a while. Additionally, a high number of occasional and one-timer users with unknown preferences and previous behaviour occur often in these domains. For this reason, modelling of user short-term behaviour become increasingly important, as it allows capturing of actual behaviour. In this thesis, we focus on the process of user modelling complexly, from identification of suitable data sources, through sessions creation from individual user actions. Consequently, we continue by the model fulfilling and by capturing chosen user characteristics. Equally important task is an application of created user model into a task, which employs the model. The main contribution of this thesis is proposal of domain and language independent method for modelling changes in user short-term behaviour on the level of individual actions in user session. The model captures the user behaviour from different time periods as well as in comparison to behaviour of other users. We evaluated the model through a task of user’s session end intent prediction. This task represents a modification of customer loss task transformed into the short-term context, where we, in online time, predict if the user will leave actual webpage or if he will continue the visit.
user model, short-term user behaviour, website data mining, behavioural changes, session end intent, multilayer modelling
To display the final thesis assignment form click on the Display the final thesis assignment form icon. The following icons - Final thesis, Thesis appendices, Supervisor's review, Opponent's review - relate to the final thesis and can be downloaded. They could be displayed on condition they have been inserted and are available publicly.
|Final thesis (final thesis appendices) unlimited|
|Reviews for final thesis unlimited|