Efficient Data Center Support: Utilizing AI to Optimize Service Requests
Monday, May 13, 2024 3:00 PM to Wednesday, May 15, 2024 4:00 PM · 2 days 1 hr. (Europe/Berlin)
Foyer D-G - 2nd floor
Research Poster
High-Performance Data AnalyticsIndustrial Use Cases of HPC, ML and QCLarge Language Models and Generative AI in HPC
Information
Poster is on display and will be presented at the poster pitch session.
Our research is dedicated to transforming customer support services in High-Performance Computing (HPC) through innovative AI-driven methodologies. Focused on analyzing ticketing data, our primary objective is to automate the categorization of HPC-related support tickets, optimizing staff allocation and response times. Beyond categorization, our research extracts comprehensive summaries from customer-agent interactions, creating a robust dataset for HPC insights and a Q&A platform for efficient interactions. Our approach extends to the integration of federated learning, fostering collaboration across centers while preserving data privacy. We aim to revolutionize CRM processes by seamlessly incorporating AI into support methodologies. Key goals include: 1. Automated Ticket Categorization: Develop an optimized method for categorizing HPC-related support tickets, streamlining workflows for improved efficiency. 2. Summarization for CRM and HPC: Generate concise summaries for HPC data within CRM systems, facilitating quicker decision-making and providing insights for Q&A purposes. 3. Q&A Chatbot: Create an interactive chatbot to handle real-time customer queries, enhancing satisfaction and reducing the workload on human agents. 4. Distributed Model Across Centers: Leverage federated learning to build a centralized language model for universal application, promoting collaboration while ensuring data security. Acknowledging challenges, our approach evolves with advanced LLM models to align customer needs with agent expertise. Currently, we're implementing LSTM for ticket categorization and LLM-based summarization, augmenting the model with historical data for effective customer question answering.
Contributors:
Our research is dedicated to transforming customer support services in High-Performance Computing (HPC) through innovative AI-driven methodologies. Focused on analyzing ticketing data, our primary objective is to automate the categorization of HPC-related support tickets, optimizing staff allocation and response times. Beyond categorization, our research extracts comprehensive summaries from customer-agent interactions, creating a robust dataset for HPC insights and a Q&A platform for efficient interactions. Our approach extends to the integration of federated learning, fostering collaboration across centers while preserving data privacy. We aim to revolutionize CRM processes by seamlessly incorporating AI into support methodologies. Key goals include: 1. Automated Ticket Categorization: Develop an optimized method for categorizing HPC-related support tickets, streamlining workflows for improved efficiency. 2. Summarization for CRM and HPC: Generate concise summaries for HPC data within CRM systems, facilitating quicker decision-making and providing insights for Q&A purposes. 3. Q&A Chatbot: Create an interactive chatbot to handle real-time customer queries, enhancing satisfaction and reducing the workload on human agents. 4. Distributed Model Across Centers: Leverage federated learning to build a centralized language model for universal application, promoting collaboration while ensuring data security. Acknowledging challenges, our approach evolves with advanced LLM models to align customer needs with agent expertise. Currently, we're implementing LSTM for ticket categorization and LLM-based summarization, augmenting the model with historical data for effective customer question answering.
Contributors:
Format
On-site
