Abstract | ||
---|---|---|
This paper proposes probabilistic models for social media mining based on the multiple attributes of social media content, bloggers, and links. The authors present a unique social media classification framework that computes the normalized document-topic matrix. After comparing the results for social media classification on real-world data, the authors find that the model outperforms the other techniques in terms of overall precision and recall. The results demonstrate that additional information contained in social media attributes can improve classification and retrieval results. |
Year | DOI | Venue |
---|---|---|
2011 | 10.4018/jitwe.2011010102 | IJITWE |
Keywords | Field | DocType |
overall precision,social media content,normalized document-topic matrix,probabilistic model,additional information,unique social media classification,social media,social media mining,multiple attribute,social media attribute,probabilistic models,social media classification,classification,latent dirichlet allocation | Data mining,Latent Dirichlet allocation,Normalization (statistics),Social media,Information retrieval,Social mining,Social media mining,Computer science,Precision and recall,Artificial intelligence,Probabilistic logic,Machine learning | Journal |
Volume | Issue | Citations |
6 | 1 | 2 |
PageRank | References | Authors |
0.40 | 23 | 1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Flora S. Tsai | 1 | 352 | 23.96 |