A Simple Approach to Case-Based Reasoning in Knowledge Bases

Das, Raj; Godbole, Ameya; Dhuliawala, Shehzaad; Zaheer, Manzil; McCallum, Andrew

doi:10.48550/arxiv.2006.14198

Cited by 2 publications

(4 citation statements)

References 16 publications

(24 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We found that 76.5% (88) indications were "reported" in the literature and 5.2% (6) were "side-effects" and 18.3% (21) as "no prior link" for TransE. For probCBR, we found 73% (84) were "reported", 6.1% (7) were "side-effects" and 20.9% (24) as "no prior link". Plausible predictions from combining TransE and probCBR can be seen in Table III.…”

Section: Select Diseases For Annotationmentioning

confidence: 74%

“…Finally, the paths obtained are applied to the query entity h q to identify the answer. When multiple answers arise, the answers are ranked by the frequency of paths that lead to them [24]. Probabilistic case based reasoning extends the original case based reasoning method by improving the identification of similar entities with k-nearest neighbors approach, and by creating a probabilistic model that estimates the likelihood a retrieved path leads to a correct answer given query relation [30].…”

Section: Knowledge Graph Completion Methods For Drug Repositioningmentioning

confidence: 99%

“…Case based reasoning (CBR) is an established technique from artificial intelligence modeled after human ability to retrieve and apply prior experience to tackling a new but similar challenge, and represents another promising approach to conduct drug repositioning [19]- [23]. While not specifically applied to drug repositioning, [24] demonstrated the utility and effectiveness of this simple approach for link prediction on knowledge graphs. Applied to drug repositioning, pathways from other similar compounds can be leveraged as evidence to identify a potential disease treatment given a compound of interest.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Drug Repurposing using consilience of Knowledge Graph Completion methods

Sinha

González

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

While link prediction methods in knowledge graphs have been increasingly utilized to locate potential associations between compounds and diseases, they suffer from lack of sufficient evidence to prove why a drug and a disease may be indicated. This is especially true for knowledge graph embedding (KGE) based methods where a drug-disease indication is linked only by information gleaned from a vector representation. Complementary pathwalking algorithms can increase the confidence of drug repositioning candidates by traversing a knowledge graph. However, these methods heavily weigh the relatedness of drugs, through their targets, pharmacology or shared diseases. Furthermore, these methods rely on arbitrarily extracted paths as evidence of a compound to disease indication and lack the ability to make predictions on rare diseases. In this paper, we evaluate seven link prediction methods on a large biomedical knowledge graph on a drug repositioning task. We follow the principle of consilience, and combine the reasoning paths and predictions provided by probabilistic case-based reasoning (probCBR) with those of a KGE method, TransE, to identify putative drug repositioning indications. Finally, we highlight the utility of our approach through two potential repurposing indications.

show abstract

Section: Select Diseases For Annotationmentioning

confidence: 74%

Section: Knowledge Graph Completion Methods For Drug Repositioningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Drug Repurposing using consilience of Knowledge Graph Completion methods

Sinha

González

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…More recently, CBR has been successfully applied in the field of knowledge-based reasoning Das et al (2020). andDas et al (2021) show that CBR can effectively learn to generate new logical reasoning chains from prior cases, to answer questions on knowledge graphs.…”

mentioning

confidence: 99%

Case-based Reasoning for Better Generalization in Textual Reinforcement Learning

Atzeni¹,

Dhuliawala²,

Murugesan³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Text-based games (TBG) have emerged as promising environments for driving research in grounded language understanding and studying problems like generalization and sample efficiency. Several deep reinforcement learning (RL) methods with varying architectures and learning schemes have been proposed for TBGs. However, these methods fail to generalize efficiently, especially under distributional shifts. In a departure from deep RL approaches, in this paper, we propose a general method inspired by case-based reasoning to train agents and generalize out of the training distribution. The case-based reasoner collects instances of positive experiences from the agent's interaction with the world in the past and later reuses the collected experiences to act efficiently. The method can be applied in conjunction with any existing on-policy neural agent in the literature for TBGs. Our experiments show that the proposed approach consistently improves existing methods, obtains good out-of-distribution generalization, and achieves new state-of-the-art results on widely used environments.

show abstract

A Simple Approach to Case-Based Reasoning in Knowledge Bases

Cited by 2 publications

References 16 publications

Drug Repurposing using consilience of Knowledge Graph Completion methods

Drug Repurposing using consilience of Knowledge Graph Completion methods

Case-based Reasoning for Better Generalization in Textual Reinforcement Learning

Contact Info

Product

Resources

About