Aquisition Data and Analytics ADA

Data Analytics

Brian B. Joseph

Deputy Director, Data Analytics

Full biography

Data Analytics: Using data to enhance acquisition outcomes


The Data Analytics division of ADA serves as the functional lead for establishing an Analytic Framework to support "Data Driven Decisions" within ASD(A). The division gathers strategic questions from the organization, integrates acquisition data from disparate DoD and Federal data stores, and conducts appropriate analytics to answer those questions. This informs acquisition strategic decision making and policy changes in OUSD (A&S) and the DoD at large, at the speed of relevance, improving acquisition outcomes. The Data Analytics Division also conducts analyses to help the Department inform the public and other parts of the U.S. Government, including the Legislative Branch, on the performance of the defense acquisition system.


Data Analytics provides cross cutting analyses of DoD acquisition by commodity type, contract, contractor, contractor incentive, portfolio, and individual program using data mining principles, analytic tools, data scientists, and functional subject matter experts.

Data Visualization

Machine Learning


Research & Reports

Working Collaboratively

Acquisition Analytic Forum

This forum creates a culture that collectively shares industry analytics best practices to improve analytics and data literacy within the acquisition workforce. It brings together OUSD(A&S) and service component representatives to discuss analytics and ensure that the correct metrics are created and used to answer the strategic questions that informs the Analytics Framework and ultimately improve outcomes.

Acquisition Analytics Board

This board is comprised of the A&S DASD’s, key members from the services, and chaired by the PDASD, AE. It is a subset of the Acquisition Visibility Steering Group. The purpose of this board is to discuss current and future analytics initiatives in A&S at the strategic level.

Training Opportunities


Advana*, DoD’s big data platform for advanced analytics, has numerous free training materials* and third party training resources for its various tools, as well as webinars and office hours*. Please reach out to the Advana Help Desk* and initiate a Help Desk Ticket to inquire about or request a training.

Advana Access Request Guide

*DoD network access only

DAU Coursera 

DAU is teaming up with the world’s largest provider of massive online open courses to extend the reach of its DoD workforce training with online DoD-Coursera programs.

Relevant courses include Data Science, covering Programming in R, Practical Machine Learning, Statistical Inference, and more.

What We Do

Data Mining

Data Mining uses the Analytic Framework to search through copious amounts of computerized data to find insightful patterns or trends.

The data mining process consists of five stages: problem definition, data gathering and preparation, matrixed collaboration, model building and evaluation, and knowledge deployment.  The problem definition stage includes strategic questions from leadership.  The data stage includes data access, sampling, and transformation. The matrixed approach consists of data scientists and SMEs working together.  The model stage includes creation, testing, evaluation, and interpretation of the model. The knowledge stage includes applying the model, custom visuals and reports, and external applications.

Covering three types of learning: reinforcement, supervised, and unsupervised, applications of machine learning as a discipline include: Identity Fraud Detection, Image Classification, Customer Retention, Diagnostics, Advertising Popularity Prediction, Weather Forecasting, Market Forecasting, Estimating Life Expectancy, Population Growth Prediction, Real-time Decisions, Robot Navigation, Learning Tasks, Skill Acquisition, Game AI, Big Data Visualization, Meaningful Compression, Structure Discovery, Feature Elicitation, Recommender Systems, Targeted Marketing, and Customer Segmentation.

Machine Learning

Machine Learning uses statistical models and algorithms to give computers the ability to do new tasks based on patterns and inference, without being explicitly programmed. This diagram shows a small sample of the applications in which machine learning is currently being used.

Tree-Based Modeling

Decision trees can be used to aid the description, categorization and generalization of a set of data, such as shown on the right with our model of total and procurement costs.

Tree-Based Model of Total Cost vs Procurement for MDAP Costs in (TY$M)

Abstract network of data points varying in size and color.

Topological Data Analysis

“Data has Shape, Shape has Meaning, Meaning drives Value.” -Gunnar Carlsson
Topological Data analysis studies the underlying shape of data by bringing together mathematics with computer science. It uses algorithms and concepts from algebraic topology to extract insights from complex multi-dimensional data structures.

Natural Language Processing & Text Mining

In machine learning, Natural Language Processing (NLP) is the computer’s ability to understand and analyze human language. Text mining is useful in analyzing patterns and unexpected trends by measuring the frequency of words, which can indicate topics of concern or importance like the diagram on the right.

Word cloud of a sample defense program, showing high frequency for the words data, program, cost, aircraft, sustainment, and production.

Get Started with DATA