Collaborative User Clustering for Short Text Streams

Authors: Shangsong Liang, Zhaochun Ren, Emine Yilmaz, Evangelos Kanoulas

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

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate our proposed method via a benchmark dataset consisting of Twitter users and their tweets. Experimental results validate the effectiveness of our proposed UCIT model that integrates both users and their collaborative interests for user clustering by short text streams.
Researcher Affiliation Academia University College London, London, United Kingdom University of Amsterdam, Amsterdam, The Netherlands
Pseudocode Yes Algorithm 1: Overview of the proposed UCIT model. ... Algorithm 2: Inference for the UCIT model at time t.
Open Source Code No The paper does not contain any statement about releasing source code or a link to a code repository.
Open Datasets Yes In order to answer our research questions, we work with a dataset collected from Twitter (Zhao et al. 2016).
Dataset Splits No Following (Gama et al. 2014), we split the dataset into two parts: half of the dataset for training, and the remaining for testing. The paper mentions training and testing splits, but does not explicitly mention a separate validation split or its size/proportion.
Hardware Specification No The paper does not provide any specific hardware details such as GPU models, CPU types, or memory used for running the experiments.
Software Dependencies No The paper mentions several topic models (LDA, DTM, TTM, To T, GSDMM) and algorithms (K-means, Gibbs sampling) but does not specify any software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow, or specific library versions).
Experiment Setup Yes For static topic models, i.e., LDA and Author T, we set α = 0.1 and β = 0.01. We set the number of topics Z = 50 and the number of clusters equal to the number of topics.