Markov-modulated Marked Poisson Processes for Check-in Data
Authors: Jiangwei Pan, Vinayak Rao, Pankaj Agarwal, Alan Gelfand
ICML 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experiments demonstrate the usefulness of our approach by comparing with various baselines on a variety of tasks. |
| Researcher Affiliation | Academia | 1 Dept. of Comp. Sci., Duke University, 2 Dept. of Statistics, Purdue University, 3 Dept. of Statistical Sci., Duke Univ. |
| Pseudocode | No | The paper describes the inference process but does not include a formal pseudocode block or algorithm listing. |
| Open Source Code | No | The paper does not contain an explicit statement or link indicating that the source code for the methodology is openly available. |
| Open Datasets | Yes | We use a dataset of Four Square check-in sequences from year 2011. Each check-in has a location (latitude and longitude) and a time stamp. The dataset was originally collected by (Gao et al., 2012a) to study location-based social networks. |
| Dataset Splits | No | The paper mentions fitting models to "500 random users" and evaluating on "test datasets", but does not explicitly provide specific training/validation/test splits (e.g., percentages, sample counts, or predefined split references) needed for reproduction. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers. |
| Experiment Setup | Yes | In our experiments, we set the number of states and topics to 50 for the national data and 30 for the Florida data. We fit all models using data from 500 random users, running the MCMC samplers for 1500 iterations and retaining the last 500 samples for the evaluation experiments described below. For every test user, we perform 500 iterations of Gibbs sampling of the user s hidden trajectory and preference vector, and use the last 200 samples for prediction. |