Air pollution is one of the major problems of the modern world. The popularization and powerful functions of smartphone applications enable people to participate in urban sensing to better know about the air problems surrounding them. Data sampling is one of the most important problems that affect the sensing performance. In this paper, we propose an Adaptive Sampling Scheme for Urban Air Quality (AS-air) through participatory sensing. Firstly, we propose to find the pattern rules of air quality according to the historical data contributed by participants based on Apriori algorithm. Based on it, we predict the on-line air quality and use it to accelerate the learning process to choose and adapt the sampling parameter based on Q-learning. The evaluation results show that AS-air provides an energy-efficient sampling strategy, which is adaptive toward the varied outside air environment with good sampling efficiency.

CA125 is a well-known tumor marker for diagnosis, monitoring, and risk stratification in ovarian cancer. It is not specific for malignant tumors and may be elevated in benign disease. In the past two decades, increasing evidence has emerged suggesting that the plasma level of CA125 can serve as a novel surrogate of heart failure (HF). CA125 in patients with HF is synthesized by serous epithelial cells in response to both mechanical and inflammatory stimuli. In patients with HF, regardless of etiology, CA125 levels correlate with the severity of clinical, hemodynamic, and echocardiographic parameters and with other biomarkers. Elevated CA125 can identify patients at high risk of rehospitalization and mortality, whether short- or long-term. Serial measurements and combination with different pathophysiology biomarkers can provide a more accurate prognosis value. It also can guide treatment as a robust biomarker of fluid overload and inflammation, particularly for diuretic dose optimization. These properties make it a very promising candidate for risk stratification and treatment guidance of HF.

Urban crowd flow prediction is very challenging for public management and planning in smart city applications. Existing work mostly focuses on spatial and temporal dependence based flow prediction that are not well suited for predictions of instantaneous flow change usually due to social emergency incidents and accidents. In this paper, we propose an Online to Offline Interaction based Dilated Casual Convolutional Neural Network framework (O2O-DCNN) for crowd flow prediction. Both online attention behavior and offline crowd shift factors are considered, in case to capture the dependence between them and make more accurate predictions especially for instantaneous flow variations. The online and offline features are processed by dilated casual convolutions and then put into CBOW model based full connected network to make interactions. The performance evaluations are based on realistic User Detail Record (UDR) dataset of a southern city in China provided by China Unicom. O2O-DCNN is compared with the other related baselines in terms of MASE and MAE. The results show that our framework is with much better accuracy, especially for instantaneous flow variation scenario.

Sensor data search is an imminent component for the booming Internet of Things (IoT). Current proposals would refer to either keyword or spatial location, then the matched sensor data should be manually selected by requesters. However, the selection is arduous as there are massive connected devices. To this end, we propose a Combinatorial-Oriented Feedback (COF) mechanism to guarantee the reliability and accessibility of feedback results via enabling an intuitive exhibition of aggregated sensor data. Two critical issues are addressed in this paper: Sensor data combination and aggregated results ranking. To facilitate the combination process, we first propose two decisive factors for aggregated result evaluation. The Multiple Sensor-data Combinatorial (MSC) problem is converted into a multi-objective optimization model. To balance the trade-off between competing quality metrics, we introduce Pareto Sensor Set (PSS) as an optimal solution for MSC problem, and devise the Elitist Directive Breeding (EDB) method to get PSS solutions. In order to speed up the search efficiency and improve the feedback recall, we then develop the Fast Elitist Directive Breeding (FEDB) algorithm, which is able to return top-k ranked results according to different searching requirements. To demonstrate the effectiveness of COF mechanism, we conduct extensive simulations based on a virtual mobile sensing scenario. The results show that our proposed COF mechanism outperforms the state-of-the-art solutions significantly in terms of search efficiency and feedback quality, and specially, FEDB responds quite faster than EDB under the command of top-k ranking.