Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Identifying vulnerabilities in trust and reputation systems
Authors: Taha D. Güneş, Long Tran-Thanh, Timothy J. Norman
IJCAI 2019 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In order to evaluate our model for identifying trust and reputation system vulnerabilities we provide a simulation environment, through which controlled experiments can be conducted for attacks focused on a target decision maker, δ. This simulation and analysis environment, along with implementations of the trust models used in this section, is freely available [Gunes et al., 2019]. |
| Researcher Affiliation | Academia | Taha D. Gunes , Long Tran-Thanh and Timothy J. Norman Electronics and Computer Science, University of Southampton, UK |
| Pseudocode | No | The paper describes methods and models but does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | This simulation and analysis environment, along with implementations of the trust models used in this section, is freely available [Gunes et al., 2019]. (Referring to [Gunes et al., 2019] which links to https://doi.org/10.5258/SOTON/D0937, containing 'src.zip') |
| Open Datasets | Yes | Observations made by witnesses are drawn from Bernoulli distributions characterising the behaviour of each provider. The parameters of these Bernoulli distributions are drawn from either a Uniform distribution or a Dirichlet with all its parameters set to 20 to produce providers that behave in a similar manner. This simulation and analysis environment...is freely available [Gunes et al., 2019]. |
| Dataset Splits | No | The paper describes running 3000 different instances for each experimental condition with dynamically generated data, but it does not specify traditional training/validation/test dataset splits. |
| Hardware Specification | No | The paper mentions 'the use of IRIDIS HPC facility at the University of Southampton' but does not provide specific hardware details such as GPU or CPU models, or memory specifications. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers, only mentioning that implementations and the simulation environment are available. |
| Experiment Setup | Yes | Parameter Value Description |P| 20 The number of provider agents |W| 20 The number of witness agents s 2 The number of witnesses under the attacker’s control t 10 The number of provider observations made by each witness. ... We consider four experimental variables: the strategy used to search for attacks (MCS or HS); the connectivity between witnesses and providers (d); the power of the attacker (ρ); and the behaviour of witnesses (θci). Other parameters are fixed as specified in Table 1. |