Food access is an unfortunate but very real problem for the many Americans that live in food deserts where the combination of distance to full service supermarkets and access to transportation makes healthy, affordable food less attainable. Today's technological innovations have the potential to address this problem, however they must be adapted to apply to the challenging socio-economic conditions of these communities. The proposed work will explore the development of heterogeneous network models, information visualization, and delivery services for addressing the problem of food access in two low-income communities in South Bend, Indiana and Detroit, Michigan. The proposed work will deeply integrate social research with technological innovation in a user-centered design-thinking framework in order to identify, understand, and meet the needs of the community stakeholders. In particular the proposed work addresses four overarching questions:What are the critical-user needs for a technology-enabled food recommender and access system?How do we model the complex food information landscape and make context-relevant recommendations for target users living under the constraints of povertyHow do we present accessible explanations of recommendations made over this complex landscape of determinants including, for example, preference, cost, and nutritional value.How feasible is a delivery hub model for addressing bridging the physical access gap?Through a series of iterative use research and design/development activities with community partners and community members, the project will develop a prototype recommendation engine that takes into account the broader context of poverty to make stakeholder-relevant recommendations for meal planning. This system will be evaluated using a mixed methods approach to understand the effects of the intervention on healthy food choice.