Novel Density-Based Clustering Algorithms for Uncertain Data

Authors: Xianchao Zhang, Han Liu, Xiaotong Zhang, Xinyue Liu

AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show the superiority of our proposed algorithms over existing ones. We use 7 UCI benchmark datasets for evaluation. ...Table 2 shows the accuracy results. ...Figure 3 shows the efficiency results (in milliseconds) on different datasets for different distributions.
Researcher Affiliation Academia Xianchao Zhang and Han Liu and Xiaotong Zhang and Xinyue Liu School of Software Technology Dalian University of Technology Dalian 116620, China xczhang@dlut.edu.cn, liu.han.dut@gmail.com, zxt.dut@hotmail.com, xyliu@dlut.edu.cn
Pseudocode Yes Algorithm 1 PDBSCAN and Algorithm 2 Expand cluster and Algorithm 3 PDBSCANi and Algorithm 4 Expand cluster max
Open Source Code No The paper does not provide any specific repository link or explicit statement about the release of source code for the described methodology.
Open Datasets Yes We use 7 UCI benchmark datasets for evaluation. The description of the datasets is shown in Table 1. These datasets are originally established as collections of data with deterministic values, we follow the method in (Gullo et al. 2008) to generate uncertainty in these datasets. and 1http://archive.ics.uci.edu/ml/
Dataset Splits No The paper uses UCI benchmark datasets but does not provide specific train/validation/test split percentages, sample counts, or explicit splitting methodology used for its experiments.
Hardware Specification No The paper does not provide any specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments.
Software Dependencies No The paper does not provide specific software dependency details, such as library or framework names with version numbers, needed to replicate the experiment.
Experiment Setup No The paper mentions parameters like Eps, Min Pts, and f value for its algorithms, stating they were 'adjusted continuously' for best performance, but does not provide specific values for these or other concrete hyperparameters (e.g., learning rate, batch size) in the experimental setup for PDBSCAN or PDBSCANi.