Approximating Model-Based ABox Revision in DL-Lite: Theory and Practice

Authors: Guilin Qi, Zhe Wang, Kewen Wang, Xuefeng Fu, Zhiqiang Zhuang

AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conducted some experiments on a data set constructed from UOBM benchmark ontology4 (see Table 1 for details of the data set). We generated ABoxes by using the UOBM generator. ... The results of our experiments are shown in Table 2.
Researcher Affiliation Academia 1 School of Computer Science and Engineering, Southeast University, China 2 State Key Lab for Novel Software Technology, Nanjing University, Nanjing, China 3 School of Information and Communication Technology, Griffith University, Australia
Pseudocode Yes Algorithm 1: At Alg(O, N)), Algorithm 2: Graph Revi(T , A, N), Algorithm 3: Search(G, A)
Open Source Code No The paper states: "We have implemented our graph-based algorithm for ABox revision in Java." and mentions storing data in a "Neo4j graph database". However, it does not provide any link or explicit statement that their implemented code is open-source or publicly available.
Open Datasets Yes We conducted some experiments on a data set constructed from UOBM benchmark ontology4 (see Table 1 for details of the data set). We generated ABoxes by using the UOBM generator. 4http://www.cs.ox.ac.uk/isg/tools/UOBMGenerator/
Dataset Splits No The paper states: "We divided each generated ABox into two parts. We used the Random class of Java to control the dividing procedure." However, it does not specify explicit training, validation, or test splits, percentages, or the methodology for how these parts were used in an experimental pipeline for reproduction.
Hardware Specification Yes All experiments have been performed on a PC with Intel Corei5-2400 3.1 GHz CPU and 6GB of RAM, running Microsoft window 7 operating system, and Java 1.7 with 6GB of heap space.
Software Dependencies No The paper states: "We have implemented our graph-based algorithm for ABox revision in Java." and mentions running on "Java 1.7". It also states, "We first transform a DL-Lite ontology into a graph and store it in a Neo4j graph database". While Java 1.7 has a version, the crucial Neo4j database is mentioned without a specific version number, which is necessary for full reproducibility.
Experiment Setup No The paper describes the process of generating data and introducing inconsistencies. However, it does not provide specific experimental setup details such as hyperparameters, learning rates, batch sizes, or model initialization settings relevant for training or evaluation.