Temporal Topic Analysis with Endogenous and Exogenous Processes
Authors: Baiyang Wang, Diego Klabjan
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The model is applied to two collections of documents to illustrate its empirical performance: online job advertisements from Direct Employers Association and journalists postings on Business Insider.com. |
| Researcher Affiliation | Academia | Department of Industrial Engineering and Management Sciences, Northwestern University, 2145 Sheridan Road, Evanston, Illinois, USA, 60208 |
| Pseudocode | No | The paper describes the model and algorithm using mathematical equations and textual explanations, but does not provide structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any explicit statements about releasing source code or links to a code repository for the described methodology. |
| Open Datasets | No | The paper states it uses "online job advertisements from Direct Employers Association" and "journalists postings on Business Insider.com". However, it does not provide concrete access information (e.g., specific links, DOIs, or citations to publicly available versions of these collected datasets). |
| Dataset Splits | No | The paper specifies training and testing dataset sizes and percentages for both case studies (e.g., 'The training data set consists of 40,449 advertisements, and the testing data set consists of 4,211 advertisements (9.4% of the sample)'). However, it does not explicitly mention a separate validation set or split. |
| Hardware Specification | No | The paper does not provide specific hardware details such as CPU/GPU models, memory specifications, or types of computing resources used for running the experiments. |
| Software Dependencies | No | The paper mentions software like "Java" and "R package stm" but does not provide specific version numbers for these or other software dependencies required to replicate the experiments. |
| Experiment Setup | Yes | We initialize the hyperparameters of LDA as follows: α = (50/K, . . . , 50/K), β = (0.01, . . . , 0.01)... For GCLDA, we let γ = 1, π0 = (1/K, . . . , 1/K), αt iid Γ(1, 1), p(η) e 0.01 |ηk|, λ Γ(1, 1), β = 0.01. We carry out the Metropolis-within-Gibbs algorithm... and run 5,000 iterations of the Markov chain with 1,000 burn-in samples... The number of topics is set to K = 50 for both data sets. |