☞ The Guiding Question ☜
Classification problems have many applications, such as in object recognition, web-page classification, and keyword extraction. This project focuses on classifying research papers taken from CiteSeerX. The goal is to create a model that automatically determines a paper’s class label. The significance of this project is that it can be extended to other tasks, such as ranking algorithms, automated key phrase detection, and collaborator suggestions within a network.
☞ The Proposed Solution ☜
The proposed model is based on Qing Lu and Lise Getoor’s supervised link-based classification method, but the model will be extended to account for citation contexts, which can increase the accuracy of the label prediction. Thus, by combining the two, we hope to achieve even more accurate classification results.