Personalised news recommendation based on a user’s browsing history

Abstract

“People may find articulating what they want hard, but they are very good at recognizing it when they see it” [35]. A significant recommendation system relies on an accurate user model to be generated. Document clustering in collaboration with the idea of user profiling are both highly regarded concepts in the field of information retrieval. A popular representation of the user’s web history documents is the vector space model, representing texts with feature vectors generated from the set of terms contained in the text. Clustering based on document-term relation matrices suffers from noise due to different words with similar meanings. The method described in this thesis uses Latent Semantic Analysis (LSA) in order to cluster the user’s documents to build the user model and in turn, find the closest related news articles that the user may deem as “interesting”. It is important to note that all of this will be done in an implicit fashion with relatively no requirements needed from the user in order to build the profile and recommend suggested articles. The Google Chrome Extension framework provides a good starting point for such a system by means of its well structured browsing history data. However, the said framework puts a condition on the system where a profile is created or updated whilst the extension is to be open. Background processing of the termdocument matrix and SVD computations are not available in this context. The clustering component and the recommendation component were tested separately using a controlled document dataset in both cases. Results from the evaluation and a user survey show that this approach is successful to separate a user’s interests by clustering the browsing history. In addition, LSA has been shown to improve the clustering results substantially. Using these clusters, accurate recommendations can be provided in an unsupervised manner, due to the separation of the user’s interests whilst using nothing but client side web technologies acting solely within the browser.