I am working on a intention oriented recommendation system, however I don't want to crawl web to get user intentions. Is there any technique to synthetically generate user intentions to prove the concept?
Could browsing behaviour be useful for this purpose? Could there be a way to generate intent from this doing an analysis of what has existed about the user over time?
Thank you for the answer, currently as you mentioned intentions are extracted by analyzing user online behavior, i.e. browsing patterns, interaction with the recommender system. What about the idea that similar users have similar preferences and hence user-based predictions can be utilized as intentions for a target user?
Dear Wajid,
Similarity between users does not indicate any intention of a target user. Intention that can be extracted may be calculated from the elapse time each user spend on reading movies reviews. This is an indication of his intention since he was interested on reading info or reviews about the movie.
Pavlos Kefalas Dear, I using MovieLens Data for experimentation in this regard, I cannot use the time taken to read the review, can you suggest any other additional information that can be incorporated that can be used as intention?
- Badges
- Science topic