We present GETL (accepted to TKDE 2026 🎉), a graph ETL framework across graph models for graph warehouses. ✨Unifies RDF, RDF-star, LPG, and relational data via a unified graph representation ⚡Supports graph transformations with Gremlin Datalog #TKDE #GraphETL #GraphModel

bioSBM: A Random Graph Model to Integrate Epigenomic Data in Chromatin Structure Prediction 1. A novel graph-based generative model, bioSBM, has been introduced to integrate epigenomic data with chromatin interaction patterns from Hi-C data. This model provides a powerful tool for understanding the relationship between chromatin architecture and biochemical processes. 2. bioSBM extends the stochastic block model (SBM) by incorporating biochemical covariates such as histone modifications and transcription factor binding. It allows for mixed memberships, enabling genomic regions to belong to multiple communities, thus capturing the context-dependent nature of chromatin interactions. 3. The model was applied to Hi-C data from the GM12878 cell line, revealing 7 distinct communities that strongly align with known biological annotations. This demonstrates bioSBM's ability to uncover latent structures in chromatin interactions beyond traditional A/B compartments. 4. bioSBM's predictive power was tested by generating Hi-C maps for unseen chromosomes and cell lines, including the HCT116 cell line under RAD21 depletion. The model showed robust performance, maintaining high accuracy across different conditions. 5. The study highlights bioSBM's potential for elucidating the relationship between biochemistry and chromatin architecture. Its interpretability and scalability make it a valuable tool for studying chromatin organization in diverse cellular contexts. 📜Paper: arxiv.org/abs/2409.14425v2 #ChromatinStructure #Epigenomics #Bioinformatics #GraphModel #PredictiveModeling

The next instalment in our series of short-form video tutorials on relationships in #SurrealDB is live! Sit back and watch our Data Evangelist, @Datastorydesign, showcase the power of graph-style relationships. 👇 #graphmodel #multiplejoins

🌐💻Google Research is making waves with its latest discovery! 📌They've unveiled Distilling Step-by-Step, an innovative approach that revolutionizes training smaller, task-specific AI models. 📌It delivers remarkable results with superior data efficiency compared to traditional approaches. 📊 📌To delve deeper into the details, check out the link below: 👇 blog.research.google/2023/09… #GoogleResearch #GraphModel #AI #LanguageModels #TechInnovation #Google #GoogleAlerts #AIInnovation #GoogleAI #TechNews

#RDF-Star/RDF* allows descriptions (e.g. scores, weights, temporal aspects, provenance, etc.) to be added to edges in a graph. RDF* extends the #RDF #graphmodel by allowing statements about statements, i.e. attaching #metadata that describes an edge: hubs.la/H0Hn7b_0

If you have trained models with Tensorflow.js, you can: - Reduce model size by up to 75% improve memory using quantization. - Optimize the prediction speed up to 20-30% (in browser) or 70-90% (in Node.js) by transforming the LayerModel to GraphModel.

brillant app — takes data from #Wikidata and builds trees. Not only genealogies, but virtually any property which is used as relationship. Try it out with people from your field! #graphmodel

The tfjs -> python model -> tflite path is the only way i am aware of. Also it only works for tf.keras models (LayersModels) only. Maybe there can be a feature request for tfjs converter to go tfjs directly to tflite. But that doesn't address the GraphModel limitation.

Read my latest publication on modelling Genomic Variants using Graphs here: ieeexplore.ieee.org/abstract… #vcf #graphs #graphmodel #model #mathematicalmodel #genome #humangenome #variants #bigdata #neo4j

We will revisit graph modeling with Northwind dataset on the upcoming webinar! Register at bit.ly/2L13DU4! #AgensGraph #graphdatabase #dataimport #graphmodel bitnine.net/blog-agens-solut…

“Graphs provide context for improved efficiency for #machinelearning algorithms because data is already connected in the #graphmodel, enabling relationships of numerous degrees of separation to be traversed and analyzed quickly at scale.” – by @amyhodler r.neo4j.com/2YXsnW5

[From the archives:] Vampire Express: Graphing a Classic '80s Choose Your Own Adventure Story // First installment in a new blog series about encountering graphs in everyday life by @joedepeau r.neo4j.com/2EkE51Q #Neo4j #GraphModel

“Graphs provide context for improved efficiency for #machinelearning algorithms because data is already connected in the #graphmodel, enabling relationships of numerous degrees of separation to be traversed and analyzed quickly at scale.” – by @amyhodler r.neo4j.com/2YXsnW5

[From the archives:] Vampire Express: Graphing a Classic '80s Choose Your Own Adventure Story // First installment in a new blog series about encountering graphs in everyday life by @joedepeau r.neo4j.com/2EkE51Q #Neo4j #GraphModel

Principal Engineer John Snelson presents how to leverage #semantic graphs in the multi-model context. #MLW19 #graphmodel #data #documentsearch

[From the archives:] Vampire Express: Graphing a Classic '80s Choose Your Own Adventure Story // First installment in a new blog series about encountering graphs in everyday life by @joedepeau r.neo4j.com/2EkE51Q #Neo4j #GraphModel

[From the archives:] Decyphering Your #GraphModel | by @idomdavis, CTO at @techmarionette
 r.neo4j.com/2RS3yVb #GraphDatabases #Cypher #Neo4j

Vampire Express: Graphing a Classic '80s Choose Your Own Adventure Story // First installment in a new blog series about encountering graphs in everyday life by @joedepeau r.neo4j.com/2EkE51Q #Neo4j #GraphModel

Learn the basics of the RDF #graphmodel and best practices for your #knowledgegraph in our upcoming training: bit.ly/2NGVUeK

Vampire Express: Graphing a Classic '80s Choose Your Own Adventure Story // First installment in a new blog series (!) about encountering graphs in everyday life by @joedepeau r.neo4j.com/2EkE51Q #Neo4j #GraphModel