Opportunistic sensing can be used to obtain data from the sensors that just happen to be present in the user's surrounding. By harnessing these opportunistic sensor con?gurations to infer activity or context, ambient intelligence environments become more robust, have improved user comfort thanks to reduced requirements on body-worn sensor deployment, and they are not limited to a conscribed location de?ned by sensors speci?cally deployed for an application We present the OPPORTUNITY Framework and Data Processing Ecosystem to recognize human activities or contexts in such opportunistic sensor con?gurations. It addresses the challenge of inferring human activities with limited guarantees about placement, nature and run-time availability of sensors. We realize this by a combination of: (i) a sensing/context framework capable of coordinating sensor recruitment according to a high level recognition goal, (ii) the corresponding dynamic instanciation of data processing elements to infer activities, (iii) a tight interaction between both in an "ecosystem" allowing to autonomously discover novel knowledge about sensor characteristics that is re-usable in subsequent recognition queries. This allows the system to operate in open-ended environments. We demonstrate OPPORTUNITY on a large scale dataset collected to exhibit the sensor-rich char- acteristics of opportunistic sensing systems. The dataset comprises 25 hours of activities of daily living, collected from 12 subjects. It contains the data of 72 sensors of 10 modalities and part 15 networked sensor systems deployed in objects, on body and in the environment. We show the mapping from a recog- nition goal to an instanciation of the recognition system. We show the knowledge acquisition and reuse of the autonomously discovered semantic meaning of an unknown new sensor, the autonomous update of the trust indicator of a sensor due to unforeseen deteriorations, and the autonomous discovery of the on-body sensor placement.