How to Apply the Skip List to the Management of a Massive Set of Replying Comments in Web Bulletins by Exploiting the Power-Law Distribution

In recent year, blogs or web discussion boards are becoming popular and widespread for netizens to express and share their opinions. In a survey conducted in 2007, more than 120,000 blogs are created in a day and more than 70 million blogs are collected by web search robots. A large number of blogs reflect current social trends and issues. In this kind of online communication, readers usually express their opinions by adding replying comments. Sometimes, the number of comments for an article can be tens of thousands and this makes it hard to manage them efficiently. The previous methods for managing database documents are not adequate for handling the dynamic nature of the replying comments in blogs. In this paper, we propose a new probabilistic method for managing comments attached on the articles of blogs. The main idea of our approach is two folds. One is to consider the probabilistic properties of the comments. The other is to reflect the dynamic changing nature of comments to the data structure. In result, we propose a modified skip list. Our simulation results show that our method is effective enough for handling the dynamic nature of replying comments and efficient considering the probabilistic distribution of comments.