This passage is a summary of Mentor Cluster. Mentor cluster, started up with Zhangyu Peng (graduate at CMU now), is developed as an importandt section of Acemap, a research engine system aiming at data mining and network visualization for scholars in all domains, especially in Computer Science.
The dataset is first partly obtained from the Internet and later predicted with our Machine Learning model based on the known mentor-student cooperation amoung publications, while this passage will mainly focuse on the visualization part, which is realized with an open source software Gephi.
The picture above is one of the mentor-cluster pictures produced with the data collected on the Internet, containing 5,000 researchers. Virious colors represent several prevalent affiliations. And the size of the node indicate the first-hierarchy students number. Those researcher in the center of the map converge into one large connected graph, while other researcher formed their own cluster revolving around.
Above is part of the mentor-cluster zoomed in. 
Stanford is represented with green in this pattern, form the picture above we can plainly tell that the students are more likely to work in the same affiliation as their mentor. 
What’s more exciting is that a huge mentor structure is uncovered, just as the picture below, the mentor cluster of William Sharpey. The students of William Sharpey reached the hierarchy of 18, over ten thousand in total. The picture above only show the former 4 highest hierarchy students, nearly 500 researcher. The size represent the hierarchy of the mentor, only in the mentor’s own cluster, uncorreated with the mentor’s fellow apprentices.
The next image is formed with data predicted through our ANN model. We extrac certain features of the known reliable relationship and uncover this relationship in the top 4 CS affiliations, CMU (Green), UCB (orange), Stanford (black) and MIT (blue).
Above is a zoom in of the mentor cluster of Prof. Fei-Fei Li
As a validation, 8 predicted students could be found at Prof. Fei-Fei Li
Below is the predicted mentor cluster from 100 authors in the fields related to Machine Learning with the highest citations. Where the depth of the mentorship tree reaches 8.
Below is the predicted mentor cluster of a certain cs area amoung affiliations of 985 in China. 
Below is the predicted mentor cluster of a certain cs area amoung affiliations of 985 in China.
This is part of the prediction result of Prof. Zhi-hua Zhou, I selected the most active students. From the map we can see that Prof. Zhou is codirecting with Prof. Jiang. 
We found it potential and exciting while bevelieving there is still a lot of work to do and info it will proffer, both for the students and researchers. If you are interested in this work or have any questoin, please contact me via wenzheng.tao@utah.edu. —
The copyright belongs to the author. For business reprint please contact the author to obtain authorization, non-commercial reprint please indicate the source.
©2017-2020 by Wenzheng Tao. All rights reserved.