Collaborative Company Profiling: Insights from an Employee’s Perspective

Authors: Hao Lin, Hengshu Zhu, Yuan Zuo, Chen Zhu, Junjie Wu, Hui Xiong

AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments were conducted on a real-world data set to validate the effectiveness of CPCTR. The results show that our method provides a comprehensive understanding of company characteristics and delivers a more effective prediction of salaries than other baselines.
Researcher Affiliation Collaboration 1Beihang University, {linhao2014, zuoyuan, wujj}@buaa.edu.cn 2Baidu Talent Intelligence Center, {zhuhengshu, zhuchen02}@baidu.com 3Rutgers University, hxiong@rutgers.edu
Pseudocode Yes ALGORITHM 1: The Generative Process of CPCTR
Open Source Code No The paper does not provide any explicit statements about releasing source code or links to a code repository.
Open Datasets No The paper mentions collecting data from 'Kanzhun2', an online employment website, and provides its URL (http://www.kanzhun.com/), but does not provide access information (e.g., a specific download link, repository, or citation for the processed dataset used in the experiments) for the specific dataset generated or used in their experiments.
Dataset Splits Yes In our experiments, we used 5-fold cross-validation. For every job position that was posted by at least 5 companies, we evenly split their job-company pairs (average rating/salary values) into 5 folds. We iteratively considered each fold to be a test set and the others to be the training set.
Hardware Specification No The paper does not provide any specific details regarding the hardware (e.g., GPU/CPU models, memory, or specific computer specifications) used for running the experiments.
Software Dependencies No The paper mentions general parameter settings (e.g., 'max iter = 500', 'αsmooth = 0.01') and references a baseline tool 'Py SVD', but does not specify version numbers for any software dependencies or libraries used in their implementation.
Experiment Setup Yes The parameter settings of different methods are stated as follows. For all methods, we set the number of latent factor to K = 5 and the maximum iterations for convergence as max iter = 500. For probabilistic topic modeling in CTR and CPCTR, we set the parameters αsmooth = 0.01. For our model CPCTR, we chose the parameters by using grid search on held out predictions. As a default setting for CPCTR, we set λu = 0.1, λb = 0.01, λv = 1000, λr = 1, λs = 1.