Perceiving Group Themes from Collective Social and Behavioral Information
Authors: Peng Cui, Tianyang Zhang, Fei Wang, Peng He
AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this paper, we explicitly validate the interplay of collective social behavioral information and group themes using a large scale real dataset of online groups, and demonstrate the possibility of perceiving group themes from collective social and behavioral information. We extensively evaluate the proposed method in a large scale real online group dataset. |
| Researcher Affiliation | Collaboration | Peng Cui, Tianyang Zhang Department of Computer Science Tsinghua University, Beijing, China Fei Wang Department of Computer Science University of Connecticut, USA Peng He Department of Social Network Operation Tencent Technology, Shenzhen, China |
| Pseudocode | Yes | Algorithm 1 REgurlarized mi XEd Regression (REXER) |
| Open Source Code | No | The paper does not provide any explicit statements about releasing source code or links to a code repository for the methodology described. |
| Open Datasets | No | The paper mentions that the dataset 'is collected from the real social network platform QQ, an MSN-style instant messenger in China' and describes its characteristics, but it does not provide concrete access information such as a link, DOI, or specific citation to a publicly available version of this dataset. |
| Dataset Splits | Yes | To evaluate the prediction performance of REXER, we randomly select 20% groups and hide their corresponding category and label set information from Y and M. After learning A and R from the remaining 80% entries, we reconstruct Y and M and calculate the loss on the hidden entries. For all the following experiments, we conduct 20-folds testing and report average results with standard deviation. |
| Hardware Specification | No | The paper does not explicitly describe the hardware used for running the experiments. There is no mention of specific CPU, GPU models, memory, or computational resources. |
| Software Dependencies | No | The paper mentions using a 'standard LDA (Latent Dirichlet Allocation) model' and 'SVM, an standard and effective multi-class classification tool', but it does not provide specific version numbers for these or any other software components, libraries, or programming languages used. |
| Experiment Setup | Yes | In the proposed REXER method, we have 3 parameters in total, including λ1, λ2 and σ. For the parameter setting, we use grid search to get the optimal parameters λ1 = 1.3, λ2 = 0.2, σ = 0.13. |