AVATAR - Automated Virtual Agent for Truth Assessments in Real-Time
University of Arizona
There are many circumstances, particularly in a border-crossing scenario, when credibility must be rapidly and accurately determined. At the same time, people constantly deceive for a variety of reasons and the process of detecting these deceptions is extremely difficult. Using artificial intelligence and sensor technologies, BORDERS researchers are developing an avatar-based screening system termed the Automated Virtual Agent for Truth Assessments in Real-Time (AVATAR). The AVATAR is designed to flag suspicious or anomalous behavior that should be investigated more closely by a trained human agent in the field. This screening technology is designed for use at Ports of Entries, airports, detention centers, visa processing, asylum requests, and personnel screening.
In YR 5, we propose to create the third-generation AVATAR and to develop the first kiosks that are field-ready. In YR 4, we demonstrated that automated agents (AVATARs) have the potential to greatly assist DHS by freeing personnel to focus on mission-critical tasks, automating interviews and document/information collection, and providing enhanced multi-sensor credibility assessments suitable for rapid screening environments.3 This was demonstrated through a successful field trial with CBP where the AVATAR conducted over one hundred interviews with Trusted Traveler program applicants--relieving existing personnel and providing detailed interview response and behavior summaries to officers to assist their final decision.
New Immigrant Survey
University of Arizona
Little is known about the origins of legal immigrants, other than their countries of birth or last residence. We currently know little about their pre-immigration labor market experiences and the ability of immigrants to translate those experiences into labor market success in this country. A key to a more coherent picture of the American migration experience is knowing why immigrants choose to migrate to the U.S. and the factors that make this decision permanent or temporary. We also need a better understanding of the factors affecting the assimilation of immigrants and their children.
This project enlists leading and promising researchers to utilize the New Immigration Survey (NIS) data in statistical and quantitative analyses of immigration. Awards will be given based on the innovativeness and quality of the proposed research and use of NIS data, including faculty awards of $30,000 each and young researcher awards (postdoctoral fellows or doctoral students) awards of $12,000 each.
Biometric Identification: Research Directions
West Virginia University
The past two decades has seen a substantial increase in biometrics activity accompanied by the deployment of biometric systems in diverse applications ranging from laptop access to border control systems. The inclusion of biometric evidence in military and criminal courts necessitates a careful examination of the scientific basis for biometric recognition. In particular, there is an urgent need to systematically review the scientific literature to determine if some of the common assumptions made about biometric traits with respect to criteria such as universality, uniqueness, permanence, measurability, performance, acceptability and circumvention, is borne out in the academic literature. Thus, the purpose of this study is to:
(a) Identify gaps in existing research and the implications on operational system risks; and
(b) Provide recommendations for further research and deployment scenarios. Pupil Dilation Characterization and Mitigation Studies
Border Patrol Checkpoint Effectiveness: Models and Metrics
University of Arizona
The Checkpoint Study project has been initiated to help Customs and Border Protection - Office of Border Patrol (CBP-OBP) assess the effectiveness of traffic checkpoint operations for the public good. The key goals of the project are to evaluate and address 1) checkpoint data integrity, consistency and accuracy, 2) measures of checkpoint impacts on local communities and 3) effectiveness metrics and models, as pointed out in the GAO Report No. GAO-09-824.
Localization and Tracking of Vehicles, Cargo and Persons
University of Minnesota, Twin Cities Campus
The overall objective of the work in this project is to develop solutions for accurate, cost effective and reliable tracking and localization of designated vehicles and cargo entering the country via the borders. The solutions developed use (to the maximum extent possible) existing technologies and are therefore cost effective. The solution developed for vehicle tracking leverages the wide-acceptance of GPS as a solution for fleet tracking. It adds to this advanced signal processing techniques to ensure that designated vehicles cannot spoof or deceive tracking efforts, thereby, providing assurance that a vehicle has (or has not) traveled along a designated route. The cargo tracking solution does the same for individual pieces of cargo (separate from the vehicle) that may be moved from one vehicle to another in the process of transit.
The work in Year 5 will focus on two objectives: The first objective will be to collect data with the GPS authenticator developed as the solution to vehicle tracking in Year 4. The data collected will be used to validate the performance of the GPS authenticator. In particular, the data will be used to determine whether the missed detection and false alarm rates of the GPS authenticator are consistent with analytical predictions. We will use data collected from tests at the University of Minnesota. We will also try to coordinate with potential DHS customers to collect field data to confirm whether the system will perform as intended in environments in which it will eventually operate.
E-Verify: Profile of Enrolled Employers
University of Arizona
Using data from the USCIS transactions database, this project proposes to examine the profile of companies using E-Verify in two states: Arizona where its use is mandated by state law, and Nevada where no such requirement exists in state law. The proposal is in two phases. Phase I is to be completed in year one of the effort and would inform implementation of Phase II in a subsequent year.
This project would work with data provided by USCIS on company E-Verify enrollment in Arizona and Nevada including company name, NAICS code, date of enrollment, and number of E-Verify screenings per month from time of enrollment to the present. These company names and NAICS codes would be cross-referenced with state and private data sources such as Dunn and Bradstreet to determine company size, verify its industry code, and profile the number and size of other employers in a given NAICS code in each state. Aggregating within NAICS codes for each state, trends in the number of monthly E-Verify screenings would be compared to trends in monthly employment data from the Bureau of Labor Statistics in order to measure the extent of their correlation.
Once the analysis is completed, the project team will meet with appropriate representatives from USCIS and other stakeholders within DHS to determine whether Phase II should be undertaken and to specifically define the questions to be addressed in Phase II.