The analysis of structured or unstructured Massive Data Sets in scientific and engineering disciplines is a problem with potential applications to Intelligence Analysis. Traditional statistics and geometry-based analysis methods are inadequate. Algorithms that make minimal assumptions on the data model and the processes that generate data are required. Further, they must handle uncertain, incomplete and noisy data and be inherently parallelizable. To address these issues, IAI has been awarded a follow-on contract entitled “Topological Robust Algorithms for Massive Data Sets via Agent-based Modular Infrastructure (TA-DA) Supporting Decentralized and Parallel Processing.” A modular component will be built that can be part of a future end-to-end system. It will distill, analyze, discover, structure and interpret relevant information hidden in massive data that is stored in distributed multi-INT databases, including but not limited to social network data, ISR sensor data, and Internet traffic data. A Topological Data Analysis (TDA) approach will be used, which recovers the topology of noisy and incomplete data points, sampled from an unknown space and embedded in a high-dimensional space. Combinatorial representations of point sets will be constructed, and algorithms will be developed for effective computation of robust topological invariants. Decentralized and parallel processing will be supported to deploy the modular system on clusters of computers. Efficient data structures will be developed, tested and validated on anonymized real-world data sets including Online Social Network (OSN) and communication networks. Cluster and cloud-based computing approaches like MapReduce in a Hadoop-based ecosystem will be used to wrap the algorithms and for implementation using open-source software. Applications include finding anomalies in social, socio-technological and communication networks and using it to deliver actionable intelligence.