Non-kinetic, asymmetric and sustainability operations all require good data for modeling human social cultural behavior (HSCB). Text communication provides good HSCB data, but it is vast, becomes obsolete quickly, and is not warehoused or mined in ways relevant to HSCB analysis. A system that can dynamically collect unfiltered textual communication data for use in HSCB modeling and analysis would be useful. The Dynamic Warehousing and Mining (DWM) framework being developed by IAI handles in-situ HSCB data collection and analysis. It is a large-scale, dynamic approach for collecting, organizing and analyzing massive data to assess HSCB dimensions for a given group, and to predict current belief states and likely intended actions. DWM has three layers: the data collection layer, the analysis layer, and the application layer. The data collection layer includes HSCB data sources, and an agent-based data collection component that automatically extracts, transforms, and loads various types of data from large-scale data sources. The analysis layer includes a HSCB feature analysis component, a data cube engine, and a data mining and modeling engine. The application layer includes a set of integrated data/model query, and visualization services. DWM utilizes software agent technology to achieve automated, scalable collection and real-time processing of HSCB data. Multiple agents work in parallel to scale up the execution of complex analysis tasks including linguistic feature analysis, data cubing, mining and modeling. DWM leverages IAI’s Agent-Based Data Miner (ABMiner) data mining platform for HSCB modeling. ABMiner integrates hundreds of data mining algorithms from IAI’s machine learning projects and open sources libraries such as Weka and RapidMiner. DWM has standardized search and query capability and can support many HSCB applications and products.