LAUSR

The expert academic personas classification are indispensable for expert intelligent automatic recommendation. It can also be applied to academic evaluation, character knowledge base construction and so on. The paper introduces a text classifier trained for the personas of expert academic classification based on support vector machine. The problem of expert academic personas is converted into text classification. First, we automatically construct text training datasets based on seed expert database from Wikipedia. Then a unified model for Chinese-English expert classification is trained after text preprocessing and vectorization. Finally, we verified the effectiveness of our method on the real dataset acquired from the Internet. Experimental results show that the proposed method outperforms the baseline methods and achieves a F1-measure of 83.5% for expert academic classification.