Open-World Knowledge Graph Completion

Authors: Baoxu Shi, Tim Weninger

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on large data sets, both old and new, show that Con Mask performs well in the open-world KGC task and even outperforms existing KGC models on the standard closed-world KGC task.
Researcher Affiliation Academia Baoxu Shi, Tim Weninger University of Notre Dame {bshi,tweninge}@nd.edu
Pseudocode No The paper describes the model components and their interactions but does not include any explicit pseudocode or algorithm blocks.
Open Source Code Yes Con Mask is implemented in Tensor Flow. The source code is available at https://github.com/bxshi/Con Mask.
Open Datasets Yes The Freebase 15K (FB15k) data set is widely used in KGC... we introduce two new data sets DBPedia50k and DBPedia500k for both open-world and closed-world KGC tasks. Statistics of all data sets are shown in Tab. 2. Also, 'we also released two new DBPedia data sets for KGC research and development.'
Dataset Splits Yes Statistics of all data sets are shown in Tab. 2. The methodology used to evaluate the open-world and closed-world KGC tasks is similar to the related work. Specifically, we randomly selected 90% of the entities in the KG and induced a KG subgraph using the selected entities, and from this reduced KG, we further removed 10% of the relationships, i.e., graph-edges, to create KGtrain.
Hardware Specification No The paper does not explicitly describe the specific hardware used for running its experiments.
Software Dependencies No The paper states 'Con Mask is implemented in Tensor Flow' but does not provide specific version numbers for TensorFlow or other software dependencies.
Experiment Setup Yes Training parameters were set empirically but without finetuning. We set the word embedding size k = 200, maximum entity content and name length kc = kn = 512. The word embeddings are from the publicly available pre-trained 200-dimensional Glo Ve embeddings (Pennington, Socher, and Manning 2014). The content masking window size km = 6, number of FCN layers kfcn = 3 where each layer has 2 convolutional layers and a BN layer with a moving average decay of 0.9 followed by a dropout with a keep probability p = 0.5. Max-pooling in each FCN layer has a pool size and stride size of 2. The mini-batch size used by Con Mask is kb = 200. We use Adam as the optimizer with a learning rate of 10-2. The target sampling set sizes for |E+| and |E | are 1 and 4 respectively. All open-world KGC models were run for at most 200 epochs.