Determining Expert Research Areas with Multi-Instance Learning of Hierarchical Multi-Label Classification Model

Authors: Tao Wu, Qifan Wang, Zhiwei Zhang, Luo Si

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments have been conducted to demonstrate the superior performance of proposed research with a real world application.
Researcher Affiliation Academia Tao Wu, Qifan Wang, Zhiwei Zhang, and Luo Si Computer Science Department, Purdue University West Lafayette, IN 47907, US {wu577, wang868, zhan1187, lsi}@purdue.edu
Pseudocode Yes Algorithm 1 EM-HM3 for Multi-instance Hierarchical Multi-label Classification
Open Source Code No The paper does not provide concrete access to source code for the methodology described, nor does it include an explicit statement about code release or a link to a repository.
Open Datasets Yes We test our algorithm with a public expertise database INDURE1. 1www.indure.org
Dataset Splits Yes Ten-fold cross validations are performed, where the regularization parameter is tuned by maximizing the sum of F1 values of all levels.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper mentions the use of 'LIBSVM' but does not provide specific version numbers for it or any other software dependencies, which are required for reproducibility.
Experiment Setup Yes Linear kernel is used in all four methods to conduct fair comparisons. For MIMLSVM, the ratio γ is set to be 20%. For EM-HM3 and HM3, we use the following loss function: ℓ(y, v) = P j cj[yj = vj] where croot = 1, cj = cpa(j)/|sibl(j)|