"As mentioned in Section 3, MAS assigns keywords, from a global set of keywords, against each paper in order to characterize it properly. For each paper, we measure how diverse its keywords are (KDI) similarly by Equation 1; here ni indicates the fraction of keywords of paper x belonging to the field i. Note that, a keyword may appear in multiple fields. For them, we consider multiple instances one for each field. (e) Topic diversity: We use the unsup"
Pseudocode:
Statistics:
Check to see if there is indeed a relationship between KDI and number of average citations.
[How specifically?]
Reference:
Chakraborty, T., Kumar, S., Goyal, P., Ganguly, N., & Mukherjee, A., (2014). Towards a Stratified Learning Approach to Predict Future Citation Counts, IEEE/ACM Joint Conference on Digital Libraries, pp. 351 - 360. doi: 10.1109/JCDL.2014.6970190
KDI
"As mentioned in Section 3, MAS assigns keywords, from a global set of keywords, against each paper in order to characterize it properly. For each paper, we measure how diverse its keywords are (KDI) similarly by Equation 1; here ni indicates the fraction of keywords of paper x belonging to the field i. Note that, a keyword may appear in multiple fields. For them, we consider multiple instances one for each field. (e) Topic diversity: We use the unsup"
Pseudocode:
Statistics:
Reference:
Chakraborty, T., Kumar, S., Goyal, P., Ganguly, N., & Mukherjee, A., (2014). Towards a Stratified Learning Approach to Predict Future Citation Counts, IEEE/ACM Joint Conference on Digital Libraries, pp. 351 - 360. doi: 10.1109/JCDL.2014.6970190