Generation and Evolution of Smart APIs (GENESIS)
December, 2016 → December, 2020Alberto Abelló, Besim Bilalli, Ayman Elserafi, Daria Glushkova, Petar Jovanovic, Rana Faisal Munir, Sergi Nadal, Oscar Romero, Shumet Tadesse, Francesc Trull, Jovan Varga
Context: Application Programming Interfaces (API) have recently experienced an unexpected growth and are shaping the way in which organizations expose data and functionalities to the outside. Companies like Salesforce.com, Expedia and Walgreens report high impacts coming from the adoption of APIs, with more than a half of their sales due to services offered by those APIs.
Problem: The creation and evolution of these APIs are still done on ad-hoc basis, with little automated support and reported deficiencies. These drawbacks hinder the productivity of developers, and impacts negatively on the time to market of those APIs and the services built on top of them.
Approach: The GENESIS project will follow a tool-supported data-driven approach to improve the automatization of the above process. In a release cycle, data coming from both APIs' usage and developers will be gathered and analysed in order to compute several indicators whose analysis will guide the planning of the next release of such APIs. This data will be also used to generate complete and accurate documentation, including non-functional requirements that will facilitate the adoption of the APIs by third parties. APIs will be organized in a semantic repository with well-defined processes around.
Domain: GENESIS will be evaluated in the domain of smart cities. This domain is specially well-suited for the data-driven approach proposed in the project, due both to the existence of huge amounts of data coming from thousands of sensors and citizens, and to the open nature of most of such data.
Interest: Siemens (Austria), TEI-Ericsson (Italy) and KPA (Israel) have expressed their interest in the results of the project.
Team: A research team composed of 10 doctors plus a working team composed of 14 PhD students at different stages of their thesis. The team brings together two research groups (i.e., GESSI and DTIM) with expertise in software engineering, data management and analytics, and conceptual modelling, in order to provide a holistic approach to the addressed problem.
ID: TIN2016-79269-R (AEI/FEDER, EU)