Arian Dhini

Classifying Purchase Decision Based on User Clickstream in E-Commerce Using Web Usage Mining

The advance of internet usage in Indonesia gives positive impact on the development of e-commerce in Indonesia where 63.5% of internet users have made online transactions. Along with e-commerce B2C growth in Indonesia, firm needs for an effective promotional strategy to understand the preferences and potential purchases for each consumer to increase return on investment (ROI). This empirical study investigated purchase decision of e-commerce users using Web Usage Mining framework. The high combination of purchasing product categories by users of e-commerce website required a multi-label classification technique that could classify those pair of purchase decision. Label Powerset method with Support Vector Machine (SVM) algorithm was applied to classify e-commerce users purchase decisions using general and detailed information features. Feature selection using Information retained 60 from 90 features. The proposed feature selection with Information Gain and parameter selection using Grid Search proved that they had an ability to enhance performance to classify purchase decision of e-commerce user. Radial Basis Function (RBF) as the most effective kernel presented an accuracy of 75.6%, with slightly difference of 2.2% with classification without using feature selection.

Classifying Indonesian Online Articles as Advertisement Placement Base Using Text Mining

Rapid development in technological aspect resulting in growing level of human needs for the latest news, so that emerged a new trend of publishing and accessing news through online media or usually called online journalism. In addition, the number of people who sell and purchase through online sites also continues to increase and this opportunity is utilized by the company and the advertiser by implementing targeted web advertising. However, the high number of articles that have been published and accessed leads to great opportunities for errors in determining where to place the ads. Therefore, it needs a system that can categorize articles accessed by users as the basis of advertisement placement by the company and this classification system can be done by applying the method of Data Mining and Text Mining. This research uses document data in the form of article content that will be categorized into twenty categories of class of advertisement by using Text Mining technique with Support Vector Machine algorithm. The results of this study may be used by companies or advertisers as a basis for placement of ads on selected online media sites.