AutoML pipeline automates trend prediction from text data

This paper introduces AutoCluster, AutoTopicModeling, and AutoTrendAnalysis, a comprehensive AutoML pipeline designed to predict emerging trends from textual data with temporal attributes. The system automates the selection of optimal clustering, topic modeling (LDA, LSA, BERTopic, NMF), and time series forecasting (Prophet, ARIMA, STL, LSTM) algorithms. By classifying topics based on forecasting accuracy, the framework identifies strong signals, weak signals, and noise, aiming to reduce manual effort and enhance prediction accuracy for users with limited machine learning expertise. Experimental results show a final RMSE of 7.099, indicating high predictive accuracy. AI

IMPACT Automates complex data analysis tasks, making advanced trend prediction accessible to users with limited ML expertise.

RANK_REASON The item is a research paper detailing a new methodology and framework for trend prediction. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• automated machine learning
• AutoRegressive Integrated Moving Average
• AutoTopicModeling
• AutoTrendAnalysis
• BERTopic
• Facebook Prophet
• Latent Dirichlet Allocation
• Latent Semantic Analysis
• long short-term memory
• Non-negative Matrix Factorization
• Seasonal-Trend decomposition using Loess

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →