Sony’s Post

📢 Excited to announce our new blog post "MuseGraph: Graph-oriented Instruction Tuning of Large Language Models for Generic Graph Mining" is now live on arXiv. In this post, we introduce MuseGraph, a novel framework integrating the strengths of GNNs and LLMs, facilitating a more effective and generic approach to graph mining. Our experimental results demonstrate significant improvements in different graph tasks, showcasing the potential of MuseGraph in enhancing the accuracy of graph-oriented downstream tasks while keeping the generation powers of LLMs. Read the full article here: https://bit.ly/4ckOoRI #graphmining #LLMs #GNNS #MuseGraph

Our recent work "Path-aware Siamese Graph Neural Network for Link Prediction" has been accepted as a Full Paper for ADMA2024. ADMA is a premier conference for data mining. #GNN #LinkPrediction #DataMining

With many using ChatGPT for open-answer analysis, Survalyzer AG presents a completely new and groundbreaking AI method that’s shown a 33% increase in results compared to ChatGPT. We will tell you all about it during our free Masterclass at the Esomar event. You can sign up here: https://lnkd.in/e2mPa33C 🔆. Starring Christian Hyka (ceo), Leander Pas and Bartosz Moszczenski. Might also interesting for you to join Kinsey, Dan, Cole, David, Diana? #survey software, #experience management, #market research, #marketing research, #Customer Experience, #CX, #Employee Experience, #EX, #consumer feedback, #employee feedback, #NPS, #brand experience, #voice of the customer, #VoC, #voice of the employee, #VoE, #Employee Engagement, #employee listening

Are Machine Learning Models Reliable? In this article, we propose a novel conformal prediction framework built on traditional machine learning models (extreme gradient boosting and random forest) to generate valid classifications of rockburst while producing a measure of confidence for its output. This paper has been accepted and published in one of the top mining journals: International Journal of Mining Science and Technology (impact factor = 11.8). Link to full text of the paper: https://lnkd.in/g8WZstcN Congratulations Ibrahim Bemah, well done!

4 scientists. 7 minutes to pitch. This is CeloScience Slam. The Celonis research community untangle some of the big challenges in Process Mining research with the Celonis Academic Alliance. Only at Celosphere. #Celosphere

Nail your ability to interpret the latent variables resulting from PCA. Nailer, the right mix of AI with ollamar and Data Mining with FactoMineR.

UMBC Professor Sanjay Purushotham is an organizer of MILETS'24, the 10th Mining and Learning from Time Series workshop, which will be held on August 25 in conjunction with KDD'24, the SIGKDD and ACM Conference on Knowledge Discovery and Data Mining. MILETS'24 focuses on synergizing research in time series analysis and mining, moving from classical methods to LLMs, discussing new and open problems, and emphasizing applications to high-impact domains. #AI #KDD https://lnkd.in/eZaggbHZ

The #Google Research, Graph Mining Team describe a series of their recent works on building more scalable graph clustering, culminating in their “TeraHAC: Hierarchical Agglomerative Clustering of Trillion-Edge Graphs” paper. They cover some of the key ideas behind this work and explain how to scale a high-quality clustering algorithm that can cluster trillion-edge graphs. Read more here 👉 https://lnkd.in/dUC4bf7s

Are you embracing AI in your day to day life? Chances you are, even if you're not proactively seeking it out. If you are attending #CIM2024 in Vancouver next week, come along to my technical talk "Questioning the Future of Mining: Why Overlook Data-Driven Transformation?" With so much data on hand there is huge potential to leverage this data to gain both insight and optimisation, with more clarity than a crystal ball. You will also see Penny Stewart & myself in booth 2125. #minedigitalization #mineengineering #mineplanning #orebodylearning #orebodyknowledge

🔍 Learn Machine Learning for Geoscientists! 🛠️ Join us on July 31st for a 2-hour online masterclass by Tom Meuzelaar from Life Cycle Geo, LLC. Dive into digital transformation in the mining industry and discover basic machine learning concepts and use cases. 📅 Register here: https://lnkd.in/dg9ffZ9m #EFGA #MachineLearning #Geoscience #DigitalTransformation #ProfessionalDevelopment