Modeling bloggers' interests

Modeling bloggers' interests becomes an important blogs research issue as it can help construct the user profile. Most researchers focus only on one feature for modeling bloggers' interests such as the combined classifier to identify interests from single posts [1], and the usage of social hypertext [2]. But, a lot of features can be used to detect the interests, such as textual, temporal and interactive features [3]. Other features, such as comments and bloggers' communities, should be used to identify bloggers' interests but they are not introduced in this post.

Two interest models are introduced, STIM (short term interest model) and LTIM (long term interest model). The first model handles interests with weak stability, and the second model handles the long period with strong stability. The two models combine textual and temporal features.

Each model depends on a window of time, a lot of posts in a certain time period, which is used in updating the original interests' vector. Forgetting function can be used to calculate the coefficients used in determining the time window [4].

References:

[1] Xiaochuan Ni, et al. “Automatic Identification of Chinese
Weblogger Interests Based on Text Classification”. In WI’ 2006.

[2] Alvin Chin, et al. “A Social Hypertext Model for Finding
Community in Blogs”. In HT’06.

[3] Chun-Yuan Teng, et al. “Detection of Bloggers’ Interests:
Using Textual, Temporal, and Interactive Features”. In WI’2006

[4] Y. Cheng, G. Qiu2, J. Bu, K. Liu2, Ye Han3, C. Wang and C. Chen. Model Bloggers’ Interests Based on Forgetting Mechanism, WWW 2008.