Blog (Bazaar)
Download as PPT, PDF1 like238 views
This blog post describes a blog recommender system that identifies a user's interests based on their social network and tags. It finds blogs for a user by comparing their tags to those of their friends and friends of friends. Blogs are then ranked based on authority, number of comments, and rank, and recommended to the user. The system was tested with 107 users, 1823 total tags, and 50 blog posts crawled for each user. Future work could improve recommendations by incorporating more data and a user's web browsing history.
1 of 10
Download to read offline
More Related Content
Viewers also liked (8)
Ad
Similar to Blog (Bazaar) (20)
Ad
Recently uploaded (20)
Blog (Bazaar)
- 1. BLOG (BAZAAR) A Blog Recommender System By: Bijaya Rath Sriti Kumar
- 2. PROBLEM STATEMENT To create a social network of bloggers and recommend them blogs according to their interests.
- 3. Recap: Identify a domain (Technocrati – web 2.0) Create a Social Network based using FOAF. Based on Data Collected design the Recommender System.
- 4. Algorithm for Recommender System Find friends of the user. Find friends of the user’s friends. Get Tags from the user and compare with the friends collected in Step 2. If more than 5 tags match select the blog. Order the blogs based on authority,rank and number of comments. Order of preference: Authority, Number of comments and Rank
- 5. Data Analysis Total User: 107 Total Tags: 1823 Number of posts pages crawled: 50 (each)
- 7. Cont..
- 8. DEMO !!
- 9. Future Work Can be more effective if we have more widespread data. Recommendation can be better if we also capture the user’s web surfing history .